update of the ecowas revised master plan for the ... · update of the ecowas revised master plan...

Update of the ECOWAS revisedmaster plan for the developmentof power generation andtransmission of electrical energyFinal Report

December 2018

Financing

11th EDF Regional Indicative ProgrammeFinancing agreement EDF/2017/ 039-384

Volume 4: Generation and TransmissionMaster Plan

TRACTEBEL ENGINEERING S.A.Boulevard Simón Bolívar 34-361000 - Brussels - BELGIUMtel. +32 2 773 99 11 - fax +32 2 773 99 [email protected]

TECHNICAL DOCUMENTOur ref.: WAPP-MP/4NT/0626321/003/03TS:Imputation: P.011966/0004

INTERNAL

Client:

Project: ECOWAS MASTER PLAN FOR THE DEVELOPMENT OF REGIONAL POWER GENERATION ANDTRANSMISSION INFRASTRUCTURE 2019-2033

Subject: VOLUME 4: Generation and Transmission Master PlanComments:

03 2019 01 15 FIN *F. Sparavier *L. Charlier *J. Dubois



REV. YY/MM/DD STAT. WRITTEN VERIFIED APPROVED VALIDATED

*This document is fully completely signed on 2019.01.15TRACTEBEL ENGINEERING S.A. - Registered office: Boulevard Simón Bolívar 34-36, 1000 Brussels - BELGIUMVAT: BE 0412 639 681 - RPM/RPR Brussels: 0412 639 681 - Bank account IBAN: BE74375100843707 - BIC/SWIFT: BBRUBEBB

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 3/176

Fina

lver

sion

ECOWAS MASTER PLAN FOR THE DEVELOPMENT OF REGIONAL POWER GENERATIONAND TRANSMISSION INFRASTRUCTURE 2019-2033VOLUME 4: Generation and Transmission Master Plan

TABLE OF CONTENT

1. INTRODUCTION .............................................................................................................. 14

1.1. Context ................................................................................................................. 14

1.2. Objectives of the project ..................................................................................... 15

1.3. Organisation of the report for the update of the ECOWAS revised master planfor the development of power generation and transmission of electrical energy .............................................................................................................................. 16

1.4. Objectives of Volume 4........................................................................................ 17

2. GENERATION MASTER PLAN ......................................................................................... 18

2.1. Introduction ......................................................................................................... 18

2.2. Methodology ........................................................................................................ 18

2.2.1. Power system modeling ............................................................................. 182.2.2. Gas network modeling ............................................................................... 212.2.3. Optimization .............................................................................................. 222.2.4. Investment and Operational constraints ..................................................... 22

2.3. Optimum short-term investment plan 2018-2022 ............................................... 23

2.3.1. The implementation of the decided projects to meet the growing demand .. 232.3.2. Towards a progressive deployment of renewable energies ........................ 262.3.3. The availability of natural gas, a challenge for the next five years ............... 282.3.4. Opportunities and challenges for a 100% interconnected network .............. 32

2.4. Optimal medium-term investment Plan 2023-2029 ............................................. 342.4.1. The exploitation of regional hydropower potential: a priority ....................... 362.4.2. Strong integration of renewable energies for an optimal energy mix ........... 382.4.3. Diversifying thermal resources to limit exposure to risk and volatility .......... 392.4.4. The interconnected network to better share the resources ......................... 40

2.5. Optimal long-term investment plan 2030-2033 ................................................... 43

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 4/176

Fina

lver

sion

2.5.1. Towards optimal exploitation of economically profitable hydroelectricresources .................................................................................................. 45

2.5.2. Towards a meshed network ....................................................................... 462.5.3. Flexibility and reliability issues on the long-term ......................................... 47

2.6. Synthesis ............................................................................................................. 48

3. TRANSMISSION MASTER PLAN ..................................................................................... 50

3.1. Methodology ........................................................................................................ 503.1.1. Static analysis ........................................................................................... 503.1.2. Dynamic Analysis ...................................................................................... 55

3.2. Short term development plan - 2022 ................................................................... 553.2.1. Towards an interconnected system ............................................................ 553.2.2. Modelling of the 2022 WAPP network ........................................................ 573.2.3. Static studies ............................................................................................. 603.2.4. Dynamic studies ........................................................................................ 663.2.5. Technical operation of the network in 2022 ................................................ 77

3.3. Mid term development plan - 2025 ...................................................................... 803.3.1. Increasing the security of the system ......................................................... 803.3.2. Modelling of the 2025 WAPP Network ....................................................... 833.3.3. Static Studies ............................................................................................ 843.3.4. Dynamic studies ........................................................................................ 873.3.5. Conclusions and recommendations for 2025.............................................. 91

3.4. Long term development plan - 2033 ................................................................... 913.4.1. Increasing the sharing of resources ........................................................... 913.4.2. Modelling of the 2033 WAPP network ........................................................ 973.4.3. Static Studies ............................................................................................ 993.4.4. Operating the network with an increasing share of renewables ................ 109

4. OPPORTUNITIES BEYOND THE BORDERS OF THE WAPP ........................................ 112

4.1. Possible interconnection with North Africa ..................................................... 1124.1.1. Technical study ....................................................................................... 1124.1.2. Economic study ....................................................................................... 1274.1.3. Conclusion .............................................................................................. 132

4.2. Interconnection opportunities with the Central African Energy Pool ............. 1324.2.1. Methodology............................................................................................ 1334.2.2. Results 1344.2.3. Conclusion .............................................................................................. 136

4.3. Connection Opportunities with Cap Vert .......................................................... 136

ANNEXE B: TRANSMISSION MASTER PLAN...................................................................... 138

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 5/176

Fina

lver

sion

National Reinforcements – 2022................................................................................... 154



Development of the dynamic model ............................................................................. 162

Small Signal Stability .................................................................................................... 166

Dynamic Security Assessment – Methodology ........................................................... 171

Dynamic Security Assessment – Results .................................................................... 173

Frequency stability ....................................................................................................... 175

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 6/176

Fina

lver

sion

TABLE OF FIGURES

Figure 1: Electrical nodes selected for the generation master plan ....................................... 19Figure 2: Distribution of the decided projects by technology at horizon 2022 (MW) ............. 24Figure 3: Distribution of selected candidates projects by technology, at horizon 2022 (MW),including the solar projects identified ..................................................................................... 25Figure 4: Energy Mix WAPP, by technology, at horizon 2022 (MW) ..................................... 25Figure 5: Expected Evolution of investment costs for solar photovoltaic projects. Source:IRENA ..................................................................................................................................... 27Figure 6: Evolution of natural gas needs in the WAPP region ............................................... 31Figure 7: Evolution of natural gas needs in the WAPP region (except Nigeria) .................... 31Figure 8: Distribution of average marginal costs by country in 2022 ..................................... 33Figure 9: Distribution of the projects decided in the medium term for WAPP per fuel type .. 34Figure 10: Distribution of the projects decided in the medium term for WAPP, per fuel type,including the potential solar projects identified ...................................................................... 35Figure 11: Energy Mix of the WAPP at then of the medium-term .......................................... 36Figure 12: Typical dispatch at the end of the medium-term (2029) ....................................... 39Figure 13: Evolution of average marginal costs by country between 2022 and 2030 ........... 41Figure 14: Distribution of the average marginal costs at 12h in 2025 ................................... 42Figure 15: Distribution of the region's average marginal costs at 21h in 2025 ...................... 43Figure 16: Distribution of long-term investments by fuel type, including the potential renewableprojects identified .................................................................................................................... 44Figure 17: Energy mix of the region at the end of the study (2033) ...................................... 45Figure 18: Average marginal costs at 12h at the end of the study (2033) ............................. 46Figure 19: Average marginal costs at 9pm at the end of the study (2033) ............................ 47Figure 20: Evolution of the installed capacity (in MW) .............Error! Bookmark not defined.Figure 21: Evolution of the installed capacity (in %) ................Error! Bookmark not defined.Figure 22: Evolution of the energy mix (in GWh) ................................................................... 49Figure 23: Evolution of the energy mix (in %) ........................................................................ 49Figure 24: Typical detail level of modelling of a country’s high voltage grid .......................... 52Figure 25: Full load flow model of the WAPP (top=West, bottom=East) ............................... 53Figure 26: Connection model of generators ........................................................................... 54Figure 27: Transmission network criticalities at short-term - 2022 ........................................ 57Figure 28: Load vs solar irradiation curve .............................................................................. 61Figure 29: Visualization of the active power flows - Peak 2022 ............................................ 63Figure 30: Non-secure N-1 contingencies - 2022 static peak scenario ................................. 64Figure 31: Results of Small Signal Stability analysis - 2022 peak. ........................................ 67Figure 32: Machine speed of Manantali (MA) and Egbin 2 (NI) following loss of one unit ofEgbin 2 – 2022 peak............................................................................................................... 68Figure 33: Results of Small Signal Stability analysis - 2022 off-peak.................................... 69Figure 34: Voltage and angle transients the following loss of NI-TB interconnection - 2022peak initial case. ..................................................................................................................... 70Figure 35: Voltage and angle transients the following loss of NI-TB interconnection, 2022 peakwith R4 and R5. ...................................................................................................................... 71Figure 36: Machine speed and angular transients following loss of MA - CIV interconnection,2022 peak with R4 and R5. .................................................................................................... 71Figure 37: Machine speed and angular transients following loss of MA - CIV interconnection,2022 peak with R4 and R5. .................................................................................................... 72

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 7/176

Fina

lver

sion

Figure 38: Machine speed and angular transients following loss of MA - CIV interconnection,2022 peak with R4, R5 and R2-A. .......................................................................................... 72Figure 39: Voltage transients along CLSG following the tripping of the Linsan (GU) –Kamakwie (SL) line – 2022 peak R4, R5 and R2-A. .............................................................. 73Figure 40: Voltage transients in BU and NR following loss of one circuit of the BU - NRinterconnection, 2022 peak with R4, R5 and R2.................................................................... 74Figure 41: System response to the loss of one unit at Akosombo at T = 50s, 2022 peak withR4, R5 and R2. ....................................................................................................................... 76Figure 42: Scheme of the interconnected WAPP system by 2022. ....................................... 78Figure 43: Median backbone connection points .................................................................... 81Figure 44: Labé - Koukoutamba line ...................................................................................... 83Figure 45: Voltage transients following loss of NI-TB interconnection with and without medianbackbone- 2025 peak. ............................................................................................................ 87Figure 46: Machine speed transients following loss of NI-TB interconnection with medianbackbone for different export levels of Nigeria- 2025 peak. .................................................. 88Figure 47: Machine speed transients with median backbone for 850 MW of import, underdifferent contingency - 2025 peak. ......................................................................................... 89Figure 48: Machine speed and angular transients following loss of MA - CIV interconnection,2025 peak. .............................................................................................................................. 90Figure 49: Proposed path of the 330 kV Western backbone ................................................. 93Figure 50: New 330 kV line Bolgatanga - Juale – Dawa ....................................................... 95Figure 51: Proposed path of the second circuit of OMVG West ............................................ 96Figure 52: Impact of inertia on the rate of change of frequency (ROCOF) [Entso-e] .......... 111Figure 53: European system equivalent and surrounding network ..................................... 113Figure 54: Proposed 400 kV AC interconnection linking the existing system (new equipmentis shown in black, the existing busses are coloured in blue) ............................................... 115Figure 55: Generator angles - Three-phase short circuit at 10s (clearance at 10.1s) -intermediate connection at NOUAKCHOTT – 0 MW power transfer ................................... 118Figure 56: Active and reactive power flow over the AC interconnection - Three-phase shortcircuit at 10s (clearance at 10.1s) - intermediate connection at NOUAKCHOTT – 0 MW powertransfer .................................................................................................................................. 118Figure 57: Generator angles - Three-phase short circuit at 10s (clearance at 10.1s) - Nointermediate connection at NOUAKCHOTT – 200 MW power transfer ............................... 119Figure 58: Active and reactive power flow over the AC interconnection - Three-phase shortcircuit at 10s (clearance at 10.1s) – No intermediate connection at NOUAKCHOTT – 200 MWpower transfer ....................................................................................................................... 119Figure 59: Generator angles - Three-phase short circuit at 10s (clearance at 10.1s) - Nointermediate connection at NOUAKCHOTT – 500 MW power transfer ............................... 120Figure 60: Active and reactive power flow over the AC interconnection - Three-phase shortcircuit at 10s (clearance at 10.1s) – No intermediate connection at NOUAKCHOTT – 500 MWpower transfer ....................................................................................................................... 120Figure 61: Proposed bipolar HVDC link ±320 kV (new equipment is shown in black, the existingbusses are coloured in blue, the connection between the converters and AC grid is notpresented in detail) ............................................................................................................... 122Figure 62: Frequency at both HVDC terminals when offering frequency support after an outageof KADUNA G at t=10s ......................................................................................................... 124Figure 63: Active power flow over the HVDC link when offering frequency support after anoutage of KADUNA G at t=10sNetwork congestions within the Moroccan power systemconsidering 1000 MW export................................................................................................ 124

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 8/176

Fina

lver

sion

Figure 64: Power flow for an export of 0 MW – No solar pv power (purple numbers representMW/MVar of loads, green numbers represent MW/MVar of generation, .../... close to thebusses represent bus voltage and angle, numbers close to the lines represent the loading in%).......................................................................................................................................... 125Figure 65: Power flow for an export of 1000 MW – 1000 MW solar pv (purple numbersrepresent MW/MVar of loads, green numbers represent MW/MVar of generation, .../... closeto the busses represent bus voltage and angle, numbers close to the lines represent theloading in %) ......................................................................................................................... 126Figure 66: Generation installed capacity Morocco in 2017, by type of fuel ......................... 127Figure 67: Generation installed capacity in Morocco in 2033, by type of fuel ..................... 128Figure 68: Schematic illustration of the cost comparison between HVDC and AC connections(source: ABB) ....................................................................................................................... 130Figure 69: Optimum exchange between North-Africa and West Africa En 2033 ............... 131Figure 70: Project to route the Inga-Calabar interconnection .............................................. 133Figure 71: ExportAtions from the CAPP to The Wapp For a fee of 40 USD/MWh .............. 134Figure 72: Exports from CAPP to WAPP for a fee of 40 USD/MWh (2033) ........................ 135Figure 73: Distribution network type load model .................................................................. 164Figure 74: Sizing incident for DSA analysis. ........................................................................ 171

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 9/176

Fina

lver

sion

TABLE OF TABLESTable 1: List of hydroelectric projects over the medium-term ................................................ 37Table 2: Investments in hydro projects in the long term ........................................................ 45Table 3: Operational limits ...................................................................................................... 51Table 4: Voltage levels modelled for each country ................................................................ 52Table 5: Decided interconnections (After 2017)..................................................................... 58Table 6: Load level - Peak 2022 ............................................................................................ 59Table 7: Load level - Off-peak 2022 ....................................................................................... 60Table 8: Country balance – Asynchronous Peak 2022 .......................................................... 61Table 9: Flows on interconnection – Asynchronous Peak 2022 ............................................ 63Table 10: List of problematic contingencies – Peak 2022 ..................................................... 64Table 11: Country balance - Off-peak 2022 ........................................................................... 65Table 12: Flows on interconnection - Off-peak 2022 ............................................................. 66Table 13: List of problematic contingencies – Off-peak 2022 ................................................ 66Table 14: List of modes with damping ration below 5% - 2022 peak .................................... 67Table 15: List of modes with damping ration below 5.5% - 2022 off-peak ............................ 69Table 16: Results of frequency stability analysis - 2022 peak with R2, R3 and R4 .............. 75Table 17: Results of frequency stability analysis - 2022 off-peak with R2, R4 and R5 ......... 77Table 18: List of recommendations to improve dynamic stability at the short term. .............. 79Table 19: Increase of TTC between Nigeria and WAPP ....................................................... 82Table 20: Load level – Peak 2025 .......................................................................................... 84Table 21: Country balance – Peak 2025 ................................................................................ 85Table 22: Flows on interconnection - Peak 2025 ................................................................... 86Table 23: List of problematic contingencies (NATIONAL) – Peak 2025 ................................ 87Table 24: Increase of TTC with the 330 kV Western backbone ............................................ 92Table 25: Load level – Peak 2033 .......................................................................................... 98Table 26: Load level – Renewable scenario 2033 ................................................................. 98Table 27: Load level – Off-peak 2033 .................................................................................... 99Table 28: Country balance – Peak 2033 .............................................................................. 100Table 29: Flows on interconnection - Peak 2033 ................................................................. 102Table 30: List of problematic contingencies (NATIONAL) – Peak 2033 .............................. 102Table 31: Country balance – Renewable Scenario 2033 .................................................... 103Table 32: Generation dispatch – Renewable Scenario 2033 .............................................. 104Table 33: Flows on interconnection - Renewable Scenario 2033 ....................................... 106Table 34: List of problematic contingencies (NATIONAL) – Renewable scenario 2033 ..... 106Table 35: Generation dispatch – Off-peak 2033 .................................................................. 106Table 36: Country balance – Off-peak Scenario 2033 ......................................................... 107Table 37: Flows on interconnection Off-peak Scenario 2033 .............................................. 109Table 38: List of problematic contingencies (NATIONAL) – Off peak 2033 ........................ 109Table 39: Generator parameters for the ENTSO-E (European) system equivalent ............ 114Table 40: Governor (IEEEG1) parameters for the ENTSO-E (European) system equivalent .............................................................................................................................................. 114Table 41: Exciter (IEEET1) parameters for the ENTSO-E (European) system equivalent . 114Table 42 : Parameters of 400 kV line ................................................................................... 116Table 43: Comparison between CSC and VSC technology ................................................ 122Table 44: Results of the study On The interconnection of the WAPP with The PEAC ....... 136Table 45: Step-down transformers parameters in distribution network type load model .... 165Table 46: List of units with PSS installed - 2022 .................................................................. 165Table 47: Complete results of the DSA analysis - 2022 ...................................................... 174Table 48: Reserve Allocation - Peak 2022 ........................................................................... 176

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 10/176

Fina

lver

sion

ACRONYMS

ADB Asian Development Bank

AFD Agence française de développement

BIO Biomass Plant

CAPEX Capital Expenditure

CAPP Central Africa Power Pool

CC Combined Cycle

CEB Communauté Electrique du Bénin

CEET Compagnie Energie Electrique du Togo

CFB Circulating Fluidized Bed

CIE Compagnie Ivoirienne d’Electricité

CI-ENERGIES Côte d’Ivoire Energies

CLSG Côte d’Ivoire – Liberia – Sierra Leone – Guinea loop

COAL Coal

COD Commercial operation Date

CSP Concentrated Solar Plant

CUE Cost of Unserved Energy

DAM with Dam

(D)DO Ordinary Diesel

DFI Development finance institutions

DI Diesel group

DNI Direct Normal Irradiation

DSO Société de distribution d’électricité (Distribution System Operator)

EAGB Electricidade e Aguas da Guine-Bissau

ECOWAS Economic Community of West African States

EDG Electricité de Guinea

EDM Electricité du Mali

EDSA Electricity Distribution Supply Authority

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 11/176

Fina

lver

sion

(E)ENS (Expected) Energy Not Served

EGTC Electricity Generation and Transmission Company

EIB European Investment Bank

ERERA Ecowas Regional Electricity Regulatory Authority

EU European Union

EUR (or €) Euro

FCFA Francs CFA

FSRU Floating Storage and Regasification Unit

GDP Gross Domestic Product

GENCO GENenration COrporation

GHI Global Horizontal Irradiation

GO Gasoil

GRIDCo Electricity Transmission Company of Ghana

GT Gas Turbine

GWh Giga Watt heure

HFO Heavy fuel oil

HRSG Heat Recovery Steam Generator

HYD Hydroelectric plant

ICC Information and Coordination Center

IEA International Energy Agency

IFI International Funding Institution

IMF International Monetary Fund

IPP Independent Power Producer

IPT Independant Power Transporter

IRENA International Renewable Energy Agency

JET Jet A1

LCO Light Crude Oil

LCOE Levelized Cost of Electricity

LEC Liberia Electricity Corporation

LFO Light Fuel Oil

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 12/176

Fina

lver

sion

LHV Low Heating Value

LNG Liquefied Natural Gas

LOLE Loss of Load Expectation

LOLP Loss of Load Probability

MMBTU Million British Thermal Unit

MMCFD Million Cubic Feet per Day

MRU Union de la Rivière Mano (Mano river Union)

N/A Not Available

NAWEC National Water and Electricity Company

NBA Niger Basin Authority

NDC National Determined Contribution

NG Natural Gas

NIGELEC Société nigérienne d'électricité

NTP Notice to proceed

O&M Operation & Maintenance

OC Open Cycle

OECD Organisation for Economic Co-operation and Development

OLTC On Load Tap Changer

OMVG Organisation de Mise en Valeur du fleuve Gambie

OMVS Organisation de Mise en Valeur du fleuve Sénégal

ONEE Office National de l’Electricité et l’Eau Potable (Morocco)

OPEX Operating Expenditure

PC Pulverized Coal

PPA Power Purchase Agreement

PPP Private Public Partnership

PSS Power System Stabilizer

pu per unit

PV Photovoltaic plant

RES Renewable Energy Sources

ROR Run of river

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 13/176

Fina

lver

sion

SAIDI System Average Interruption Duration Index : Indicateur de la duréemoyenne de coupures sur le système

SAIFI System Average Interruption Frequency Index : Indicateur de lafréquence moyenne de coupures sur le système

SBEE Société Béninoise d'Energie Electrique

SENELEC Société nationale d'électricité du Sénégal

SOGEM Société de Gestion de l'Energie de Manantali

SONABEL Société nationale d'électricité du Burkina

ST Steam Turbine

SV (or VS) Standard Value

SVC Static Var Compensation

TCN Transmission Company of Nigeria

TSO Transmission System Operator

USD (or US$ or $) US Dollar

VRA Volta River Authority

WAGP(A) Western Africa Gas Pipeline (Association)

WAPP West Africa Power Pool

WT Wind Farm

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 14/176

Fina

lver

sion

1. INTRODUCTION

1.1. Context

The Economic Community of West African States (ECOWAS) is a regionalcommunity with a surface of 5.1 million of square km which represents about 17%of the African continent. With a population of more than 300 million inhabitants in2017, ECOWAS Member States are home to about one-third of the population ofsub-Saharan Africa.

ECOWAS has been created with a mandate of promoting economic integration inall fields of activity of the constituting countries. The fifteen-member countriesmaking up ECOWAS are Benin, Burkina Faso, Cape Verde, Cote d’Ivoire, TheGambia, Ghana, Guinea, Guinea Bissau, Liberia, Mali, Niger, Nigeria, SierraLeone, Senegal and Togo. The ECOWAS treaty (also known as treaty of Lagos)established the Community during its signature in Lagos (Nigeria) on May 28th,1975.

One of the most important steps of economic integration in the field of energy wasthe creation, in 2006 of the Western African Power Pool (WAPP). The WAPPpromotes the integration of the national power systems of the fourteen inlandcountries into a unified regional electricity market with the ultimate goal ofproviding, in the medium and long-term, a regular and reliable energy atcompetitive cost to the citizenry of the ECOWAS region

However, the region, which is characterized by a great diversity in terms of culture,language, demography and resources, faces enormous challenges in providingaccess to sustainable energy for its population. But the 15 ECOWAS MemberStates are driven by a common desire to offer “affordable, reliable, sustainableand modern energy for all”, as per the three main goals of the Sustainable Energyfor All (SE4All) initiative, launched by the United Nations Secretary-General.

West-African countries have a great opportunity to reach their objectives thanksto the vast untapped potential in renewable energy (including solar, wind,bioenergy and hydro-power). The Energy Transformation will happen both on-gridand off-grid. It involves the development of mini-grids with hybrid powergeneration, centralized and decentralized renewable projects potentially coupledwith a more flexible demand side, enabled by storage and smart-meteringtechnologies.

Several initiatives like the African Renewable Energy Initiative and the ECOWASpolicy on Renewable Energy support this transformation. However, such arevolution requires financing, leadership and international cooperation. In thiscontext the West African Power Pool is playing a significant role by supporting thedevelopment of major energy projects in the region.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 15/176

Fina

lver

sion

1.2. Objectives of the project

The West African Power Pool promotes cooperation and supports thedevelopment of regional projects. In 2012, the Authority of the ECOWAS Headsof State and Government approved, through Supplementary Act A/SA.12/02/12,a list of 59 Priority Projects for the subregion that emanated from the update ofthe ECOWAS Revised Master Plan for the Generation and Transmission ofElectrical Energy prepared by Tractebel. .

Considering the evolution of the energy landscape,the socio-economic context ofWest Africa over the last 5 years and the difficulty in mobilizing public andconcessional financing in the sub-region, the development of the power systemin West Africa deviated from what was foreseen in 2011. A lot of challenges affectthe utilities efficiency on several aspects including financial, regulatory, technicaland organizational points of view.

Another key parameter which should affect the energy development roadmap ofWAPP region is the expected increase penetration of Renewable Energy Sources(RES). Thanks to the significant decrease of costs and increased willingness forthe transition to sustainable energy, many WAPP countries have revised theirRES targets and launched RES projects.

Consequently, while some flagship generation and transmission projects weredeveloped in the region, some of them are still under development or werestrongly delayed while, in parallel new non-anticipated projects emerged.

In this context, the study presents four different main objectives:

· Assessing the implementation status of the priority projects identified in2011, understanding the main challenges and barriers to the development ofthese projects and identifying the lessons learned that will be taken intoaccount when updating the Master Plan;

· Identifying the main challenges and critical factors affecting theperformance of utilities in their activities as a public service and proposing anew action plan and mitigation measures to address these constraints in along-term perspective;

· Assessing the opportunities and constraints for the deployment of RenewableEnergy Sources in the sub-regional power system (potential, economics, gridconstraints…);

· Presenting a clear, comprehensive and coherent view of the futuredevelopment of power generation and transmission facilities with a list ofpriority projects for West Africa that takes into account the new drivers ofelectricity generation and consumption, while integrating the currentdevelopment of the power system at national and regional level and whileproviding recommendations for facilitating the implementation of the projects.

This will lead to an update of the ECOWAS Master Plan for Generation andTransmission of Electrical Energy, a comprehensive study providing a rationalbasis for decision making and implementation in the power sector.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 16/176

Fina

lver

sion

1.3. Organisation of the report for the update of theECOWAS revised master plan for thedevelopment of power generation andtransmission of electrical energy

The report is divided into five main volumes corresponding to the five maindeliverables of the study.

VOLUME 1: Executive Summary

Volume 1 is the synthesis of the Final Report of the update of the revisedECOWAS Master Plan. It contains the main recommendations of the studyconcerning the future development of the electricity generation and transmissioninfrastructures as well as a list of priority projects and the implementation strategyof these projects.

VOLUME 2: State of play of the current situation of the electricity systemand perspectives

Volume 2 consists of a synthesis of data collected and assumptions used in thecontext of this project, and in particular for the update of the generation andtransmission master plan.

VOLUME 3: Challenges and Action Plans for electricity Companies

Volume 3 aims at presenting the main challenges and critical factors affecting theperformance and the sustainability of utilities members of WAPP and atrecommending a new action plan and mitigation measures to address thesecritical factors from a transversal perspective...

VOLUME 4: Generation and Transmission Master Plan

Volume 4 is devoted to the results of the generation and transmission master plan:It presents a robust and economically optimal development plan while taking intoaccount the current state of the energy sector in West Africa and opportunities fordeveloping renewable energy sources in the region while ensuring the technicalstability of the interconnected system

VOLUME 5: Priority Investment Program and Implementation Strategy

Volume 5 focuses first on carrying out a review of the implementation of theECOWAS 2012-2025 Master Plan and assessing the causes of the gaps betweenwhat was initially planned and what was concretely achieved, allowing someeffects to be taken into consideration for the development of the 2017-2033updated master plan. Then, a new list of priority investment projects is drawn upon the basis of the generation-transmission master plan and a strategy isrecommended for the progressive implementation of these projects.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 17/176

Fina

lver

sion

1.4. Objectives of Volume 4

This volume is dedicated to the results of the development phase of the electricitysector and aims to present the optimal generation and transmission master planfor West Africa.

The objective of this master plan is to find the combined optimum between thedevelopment of generation facilities on a regional scale and the development ofthe intra-regional transmission system to allow the supply of electricity reliably andat a lower cost. This optimization shall take into account from a technological pointof view the classification of renewable and hydroelectric resources, the optimumthermal technologies for the region and appropriate interconnection standards. Itshall rely on the existing regional, sub-regional and national generation masterplans. It shall also take into account the emergency plans identified at the regionallevel or at the level of each country. This generation master plan has also beenaccurately verified by evaluating the static and dynamic performance of the overallsystem (generation and transmission) to ensure optimal operation of theinterconnected system.

Note that the master plan focuses on the West African system. Nevertheless, forthe sake of completeness, the impact of a WAPP connection with other powerpools is also mentioned in this report:

· From the technical and economic point of view for a potential interconnectionwith Morocco via Mauritania;

· From the economic point of view for a potential interconnection with the CentralAfrican Power Pool.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 18/176

Fina

lver

sion

2. GENERATION MASTER PLAN

2.1. Introduction

The generation master plan corresponds to the optimal investment plan in thedifferent units of generation on the short, medium and long term.

This master plan is derived from a complex optimization, the aim of which is todetermine the optimal investments to be achieved in order to obtain the systemwith the lowest discounted costs.

At the level of generation, the optimization focuses solely on the selection of thecandidate units, which are currently under study, or standard units proposed bythe consultant. Existing and decided units are indeed part of the master plan in amandatory manner.

Regarding the presentation of the results of the master plan done in this report,the approach chosen here is to highlight the major trends that emerge in the shortterm (2018-2022), the medium term (2022-2029) and the long term (beyond2030). The aim of this approach is to allow the readers of this report to be able tohave a direct overview of the optimum evolution of the region's generationcapacity.

The master plan presented below focuses on the reference scenario, in which nointerconnection with other non-ECOWAS countries is considered. Then, theimpacts of possible interconnections with Morocco or PEAC are analyzed indedicated sections further in the report.

2.2. MethodologyThe establishment of a generation-transport master plan is based on thedevelopment of a mathematical model representing the region's energy system inan adapted software.

The software used in this study is PRELE. The latter, developed by Tractebel, isdedicated to long-term system planning and therefore aims to determine theinvestments and operating conditions of the system in such a way as to minimizethe overall cost of the system.

2.2.1. Power system modelingThe West African power grid was thus modelled in PRELE, in the form of variouselectrical nodes connected to each other by means of transport lines.

2.2.1.1. ELECTRIC NODE

Each electrical node represents a geographical area which comprises theelectrical load as well as the generation available. The appropriate choice of thenumber of electrical nodes results from a compromise between the increasingcomplexity with the number of nodes and the level of detail required.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 19/176

Fina

lver

sion

For example, some member states whose network size is relatively small arerepresented in the form of a single node: This is the case of The Gambia, Guinea-Bissau, Liberia, and Sierra Leone.

Other member states with larger networks were modelled using several nodes. Inparticular, two nodes were used to represent Benin (north and south), BurkinaFaso (Ouagadougou and Bobo-Dioulasso), Côte d'Ivoire (north and south),Ghana (north and south), Guinea (north and southeast), Mali (Bamako andSikasso), Niger (Niamey and north), Senegal (Dakar and Tambacounda) andTogo (North and south). Nigeria is separated into three different nodes (south,north and east).

Figure 1: Electrical nodes selected for the generation master plan

For each of these nodes, the evolution of the load as well as of the generation arefilled in the model.

2.2.1.2. DEMAND MODELING

The demand is modelled in PRELE using a typical daily load curve for eachmember state, whose peak load evolves on the horizon considered according tothe forecast of the demand made in the data collection report.

When a member state is made up of more than one node, a pro-rata1 was madeon the total demand to spread it on these different nodes.

1 Made on the basis of the demand for the different geographic areas

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 20/176

Fina

lver

sion

2.2.1.3. MODELING OF GENERATION UNITS

The generation units are, for their part, distributed within the different nodes of thePRELE model, taking up their main characteristics, namely:

· Available power· Technology· Fuel consumption· Downtime for maintenance· Downtime due to accidental failure· Investment costs· Operational costs· Date of commissioning· ...

At this level, the existing units, which are part of the system as of the starting year,should be distinguished from the project units that can be integrated into thesystem in future years. As a reminder, the generation units in the project werethemselves classified during the data collection phase in projects decided orcandidates according to the following criteria:

· Decided units: units whose construction is underway or has been decided fora specific date of commissioning (completed studies and insured financing)

· Candidate units: units for which the studies are not yet completed or for whichfunding has not yet been found

The decided units must be incorporated into the investment plan, taking intoaccount their date of commissioning. The candidate units, for their part, can beselected by PRELE to enlarge the existing generation capacity, if it makes senseeconomically, from a given date of commissioning.

It should still be mentioned that PRELE may also decide to invest in standardgeneration units, which are not part of the lists of projects collected from MemberStates, but which may reveal interesting on the techno-economic level.

2.2.1.4. RENEWABLE ENERGY GENERATION

The renewable energy generation units considered in the study are:

· Hydroelectric power plants· Solar photovoltaic power plants· Wind turbines

For each of these technologies, a generation curve is considered in theoptimization according to:

· Site and project characteristics in the case of hydroelectric power plants;· The geographical location for photovoltaic solar power plants and wind

turbines. In particular, the following curves are taken in consideration:- Solar curves representing the evolution of solar irradiation during the 24

hours of the day in the different geographical areas of the study- Wind curves representing wind speed during the 24 hours of the day in the

different geographical areas of the study

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 21/176

Fina

lver

sion

2.2.1.5. TRANSMISSION LINE

The different electrical nodes are connected by transmission lines. Again, thesame approach is used for the lines existing, decided and candidate. PRELEcan therefore decide to invest in a candidate transmission line if this investmentlowers the total cost of the system.

These lines are inserted into the model by mentioning their main characteristics:

· Transfer capacity· Length· Losses (per unit)· Investment costs· Voltage· Date of commissioning

2.2.2. Gas network modelingIn parallel with the electric model, PRELE allows the integration of a gas network,in order to realistically model the generation of gas units.

Similarly to the power grid, the gas network is based on the existence of nodesthat may be connected by pipelines.

Under this master plan, the following gas network was considered:

· Gas nodes:- The following countries were considered to be gas producers, given their

existing gas resources :§ Nigeria§ Ghana§ Côte d’Ivoire§ Senegal

- Beyond 2025, it becomes possible to invest in LNG-type gas projects("Liquefied natural gas") in the following countries:§ Ghana§ Côte d’Ivoire§ Senegal§ Benin§ Togo

- Finally, the consumer gas nodes cover the following countries:§ Nigeria§ Ghana§ Côte d’Ivoire§ Senegal§ Benin§ Togo§ Pipeline : The only pipeline that was considered in the study is the

WAGP (West African Gas pipeline) from Nigeria and linking Benin, Togoand Ghana.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 22/176

Fina

lver

sion

The resources associated with each node and pipeline are those mentioned inthe volume 2 of this document.

2.2.3. OptimizationThe PRELE model described earlier forms a linear program under constraintswhose objective function is to minimize the current total cost of investments andoperations.

The main results of the optimization are, for each year of the planning period:

· The installed power of the generation units to be installed at each node withtheir investment costs;

· The transfer capacities of the different transmission lines to be installedbetween the different nodes with their investment costs;

· The energy produced by each generation unit with their generation costs;· The power provided by each unit at the different hours of the day and the power

transmitted each hour on the different lines;· The depletion of gas resources at each node and the associated cost of gas

consumption· The quantities of gas passing through the pipeline

These results serve as a basis for the development of the generation-transmissionmaster plan that is proposed in this document.

2.2.4. Investment and Operational constraintsSome additional constraints have been introduced in the optimization regardingthe investment opportunities in different types of generation units, in order to makethe investment programme more realistic.

2.2.4.1. CONSTRAINTS INVESTMENTS FOR COMBINED CYCLES IN NIGERIA

It was also decided to impose in Nigeria an annual investment constraint in thecombined cycle units of up to 1000 MW per year2, this limit rising to 1500 MW peryear from 2030. The goal of this constraint is to limit the investments in thisstandard technology in order to have a realistic investment plan.

2.2.4.2. CONSTRAINTS PROJECT CANDIDATE COAL

With regard to coal projects, the reference scenario of the generation master planwhich is presented below considered only the decided coal projects, leaving asidethe investment opportunities in the candidate units.

This choice was made in order to take into account the reluctance of the variousfunding partners against this technology, given the negative impact of the latteron the environment.

2 This limit being taken from the last master plan of Nigeria

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 23/176

Fina

lver

sion

2.2.4.3. TECHNICAL MINIMA

Technical minima were also considered in the optimization for the following units:

· Hydroelectric power plants: minimum 30% for irrigation reasons· Combined cycles: minimum 40% for technical reasons· Coal units: minimum 40% for technical reasons

2.3. Optimum short-term investment plan 2018-2022

2.3.1. The implementation of the decided projects to meet thegrowing demandThis first period of the master plan is naturally dominated by the commissioningof decided projects. These account for a total of 8 386 MW of installed powerwhose distribution by technology is shown in the figure below. Natural gasinstallations represent the bulk of these decided investments (53%, i.e. 4 455MW). The availability of gas will therefore be a major issue for the next five yearsand will have to be ensured to guarantee the viability of this master plan.

Most of these decided gas units are developped in Nigeria (Azura 450 MW, OkpaiII 300 MW and AFAM III 240 MW) and in Ghana and Côte d'Ivoire (Cenpower 360MW, Rotan 330 MW, Amandi 240 MW in Ghana; Ciprel V 412 MW and Azito 253MW in Côte d'Ivoire).

Many hydroelectric power plants are also planned in the short-term, particularlyin Guinea (Souapiti 450 MW, Fomi 90 MW, Kogbedou 58 MW and Frankonedou22 MW), but also in Mali (Gouina 140 MW), Niger (Kandadji 130 MW), TheSenegal (Sambangalou 128 MW), Côte d'Ivoire (Gribo-Popoli 112 MW, Singrobo44 MW) and Nigeria (Zungeru 700 MW, Kashimbilla 40 MW).

At the level of the photovoltaic technology, most of the projects decided aredeveloped in Niger (210 MW), then in Burkina Faso (105 MW), in Ghana (102MW), in Côte d'Ivoire (100 MW) and Mali (50 MW).

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 24/176

Fina

lver

sion

Figure 2: Distribution of the decided projects by technology at horizon 2022 (MW)

In addition to these projects, 1633 MW of new thermal and hydro units arenecessary to meet the demand. Therefore, the most important candidate projectsthat emerge from short-term optimization are the combined cycle of Egbin 2+ (firstphase of 1200 MW) in Nigeria, as well as the hydroelectric power unit ofBoutoubre (156 MW) in Côte d’Ivoire.

Alongside those investments, the consultant has identified potential PV solarprojects up to 2602 MW that could be developed on this short term period, in orderto reduce the energy costs in the), given the important decline of prices for thistechnology on the horizon considered3.

The volume of potential projects depends on the solar potential of the region butalso on the limits of investment and exploitation, which are proportional to thedemand.

3 It should be noted that the decline in wind prices is not yet significant enough toinvest in this technology in the short term.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 25/176

Fina

lver

sion

Figure 3: Distribution of selected candidates projects by technology, at horizon 2022 (MW), including the solar projectsidentified

Finally, the optimum energy mix which results from the optimization is representedon the Figure 4: Energy Mix WAPP, by technology, at horizon 2022 (MW) below.We observe that PV technology represents 4% of the annual energy generation,given the intermittent nature of the resource. It appears also very clearly thatnatural gas continues to play a major role in the energy supply of the sub-region.

The detailed list of the invested projects by State-Member can be found inappendix A.

Figure 4: Energy Mix WAPP, by technology, at horizon 2022 (MW)

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 26/176

Fina

lver

sion

2.3.2. Towards a progressive deployment of renewable energiesAs it was already mentionned above, it is recommended to increase the share ofrenewable in the generation master plan for the region on the short-term horizon.

Therefore, the hydroelectric plants should play a more and more important role inthe conventional energy mix of the subregion. Moreover, an important volume ofpotential photovoltaïc plants were identified, such that this technology shouldalready occupy a significative place in the generation capacity of the region, incomplement with conventional thermal and hydroelectric decided units.

Wind energy and biomass should remain marginal on the short-term, given thecost structure and the limited potential in the region.

2.3.2.1. DEVELOPMENT OF HYDROELECTRIC POWER PLANTS

This short-term horizon is characterized by the commissioning of manyhydroelectric generation units totalling 2103 MW. Of these, most are decided units(1947 MW) and only Boutoubre (156 MW) is chosen from the candidate projectsby the optimization.

However, most hydro projects will be commissioned on the medium term, asdetailed below. This is explained by the duration of construction of this kind oflarge-scale projects that are often of the order of 4 years, exceeding the short-term horizon as defined in this study.

2.3.2.2. DEVELOPMENT OF PHOTOVOLTAIC SOLAR POWER PLANTS

The penetration of photovoltaic solar power plants into the region is due to severalfactors.

The first is attributable to the rising costs associated with thermal power plants,particularly in view of the expected increases in the price of the various fuel oilsused (see Volume 2).

The second is related to the saturation of hydro projects that amount to 2100 MWon the short-term horizon.

The third factor is the significant fall in prices expected for solar photovoltaictechnology, as is recalled at the figure below. Indeed, while the average cost of asolar project at the beginning of the study is 1500 USD/kW, this falls to only 1000USD/kW in 2022, a reduction of 33% in 5 years.

Nevertheless, despite this significant fall in prices, the development of thistechnology on a large scale is generally justified only in the north of the region onthe short-term, given that the conditions of irradiation in the south are not enoughto make significant investments on the horizon up to 2022.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 27/176

Fina

lver

sion

Figure 5: Expected Evolution of investment costs for solar photovoltaic projects. Source: IRENA

Of the 3 457 MW of proposed solar projects, 855 MW are decided, 1352 MW arecandidate projects already identified by the state members and chosen by theoptimization, and 1250 MW are potential additional standard projects. The list ofthese potential standard projects should vary from 50 to 250 MW, depending onthe location of the project, the local demand, the capacity to export the power andthe availability of land.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 28/176

Fina

lver

sion

2.3.2.3. MARGINAL DEVELOPMENT OF WIND ENERGY

In addition to the decided projects (150 MW in Senegal), the economic analysisdoes not recommend the development of wind turbine project as a short-termpriority. Indeed, given the relatively low wind potential of the subregion and thecost of technology, the other alternatives of solar energy and hydropower appearmore interesting from an economic point of view.

Regarding the costs, wind turbines are recognized as a mature technology andtrends in cost reduction are less important than those observed for photovoltaic.Nevertheless, scale effects will allow a gradual reduction of the cost for windturbines, which, combined with the saturation of other resources, could pave theway for this technology in the longer term.

2.3.2.4. INVESTMENTS IN LINE WITH THE TARGETS SET FOR RENEWABLEENERGY

The energy mix 2022 of the region shows that the renewable generation canpotentially reach up to 29% (25% from hydro and 4% from solar), which is slightlybelow the ECOWAS 2020 objective (35% renewable generation including largehydro). This slight delay is due to the backlog in recent years in theimplementation of the projects. This delay should be gradually absorbed, inparticular through the acquired experience of member states in monitoring suchprojects.

2.3.3. The availability of natural gas, a challenge for the next fiveyearsAs already noted above, investments in gas plants represent the most importantpart of the investments with 5 780 MW to be installed by 2022 in the whole region,i.e. 46% of the total investment considered.

With these new investments, the share of technologies using the natural gas inthe energy mix rises to 64% in 2022. It appears hence clearly that gas availabilitywill play a crucial role in ensuring the viability of the management plan presentedin this report.

Lower risk approach to meet gas demand

Given the importance of the gas resource for the reliable supply of electricity inWest africa, the impact of the unavailability of this resource on the results of themaster plan and the variability in the cost of gas was investigated.

With regard to the availability of the resource, the dependence of the subregionon a single source of supply creates a major risk for countries.

The majority of these needs are obviously concentrated in Nigeria, accounting for77% of the region's gas consumption on average on the study horizon.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 29/176

Fina

lver

sion

In addition to Nigeria accounting for 77% of the region's gas consumption onaverage on the study horizon, the countries for which gas availability is also crucialare Ghana, Côte d’Ivoire and Senegal. Ghana and Côte d'ivoire are characterisedby indigenous reserves that are decreasing over time. These countries willtherefore have to guarantee the security of supply via other gas sources, whetherLNG units or WAGP. The exploitation of gas resources in Senegal is planned tostart in 2025 and gas requirements are expected to grow up to 183 Mmscfd onthe horizon 2033.

In Benin, the development of combined-cycle gas power plant projects, includingthe 450 MW regional WAPP project, also calls for the development of gas supplyinfrastructures. The proposal made in this master plan is to enhance the reliabilityof the WAGP, providing for gas supply opportunities not only from Nigeria but alsofrom Ghana. Togo could also benefit from such a development of the gas network.

In conclusion, it is advisable to diversify the sources of supply: indigenous sourcesin Nigeria, Ghana, Côte d'ivoire and, in the medium term, in Senegal, to which areadded gas sources imported via LNG terminals recommended in Côte d’Ivoireand Ghana, or via the WAGP.

Finally, from the point of view of the cost of gas, the analyses conducted showedthat a variation of this factor did not significantly alter the optimal investment planon the horizon of the study. A slight slippage of renewable projects (hydro andsolar PV) is nevertheless observed during the period. However, given that thecost of natural gas accounts for approximately 45% of the total cost (investment+ operation) of the master plan over the study period, any change in the cost ofthe resource will affect the total cost of operations in a significant way.

2.3.3.1. THE IMPORTANCE OF SECURING NATURAL GAS SUPPLY

Given that the possibilities of investment in LNG projects start only in 2022, theavailability of gas over the period will depend largely on the reliability of thecountry-specific resources (Nigeria, Ghana, Côte d'ivoire and Senegal), as wellas of the West Africa Gas Pipeline.

In the absence of gas resources, countries would be forced to exploit the powerplants with heavy fuels or, in the worst case, to halt the operation of thermal powerplants, which would have negative consequences for the economy andpopulation.

Therefore, the short-term priority objective lies in securing the gas supply.

2.3.3.2. INVESTMENTS IN LNG TERMINALS FROM 2022

As presented in the methodology section, the master plan considers investmentin LNG units starting from 2022. The installation of such infrastructure musthowever be justified economically given the investment costs as well as the highercost of gas for these facilities.

Nevertheless, despite these relatively large costs, the optimization indicates thatit is interesting to invest in LNG units from 2022 in Ghana and Côte d’Ivoire, inorder to be able to supply combined cycles that are not running fully due to thelack of gas available. As a reminder, Ghana and Côte d’Ivoire will actually have2040 MW and 1468 MW of combined cycles in 2022, respectively.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 30/176

Fina

lver

sion

The estimated needs in MW thermal (MWth) of LNG for Ghana and Côte d'Ivoireon the horizon of the study (in 2033) are respectively of 3500 Mwth and 2500Mwth.

It is recommended to install a first phase of 1000 Mwth of these units from 2022in these two countries in order to be able to fully utilize the combined cycle units.

It should still be noted that it is recommended that the terminal LNG of Ghanaserve also to feed Benin via the West Africa Gas Pipeline, in order to supplementthe insufficient supply from Nigeria, given the relatively large size of combined-cycle projects in Benin (BID, Maria Gleta regional project).

2.3.3.3. THE IMPORTANCE OF INVESTING IN COMBINED CYCLES

As a reminder, a combined-cycle power plants consists of one (or several) gasturbine turbines combined with a steam turbine. The operating principle is basedon the use of the heat from the heat of the exhaust fumes out of the gas turbinesto produce steam which is then relaxed in the steam turbine.

This technique allows to achieve efficiency up to 62% with the current technology,which is more important than the efficiency associated with the use of a gasturbine alone (known as open-cycle) that is around 34%.

These combined-cycle power plants are nevertheless less flexible than open-cycle power plants. However, if one considers in this master plan that these plantsare mainly intended to run in base, it is more economically interesting to invest incombined cycle units.

This is verified in the optimization as the two candidate projects of gas-fired powerplants that emerge are the combined-cycle plants of Egbin In Nigeria and MariaGleta (regional project of the WAPP) in Benin.

According to this principle, it will also be interesting to convert the many open-cycle plants of Nigeria into combined cycles in the medium and long term.

2.3.3.4. THE GAS TO REPLACE THE HFO AND THE DDO

Finally, one can also note that the increase in the use of gas in the energy mix ismainly at the expense of the consumption of HFO and DDO which only contributesup to 5% in 2022 (whereas they still accounted for 14% in 2017).

However, in the absence of a possible short-term alternative, countries in thewestern part of the region (Senegal, The Gambia, Guinea-Bissau, Sierra Leoneand Mali) still rely on these heavy fuels until 2022 to ensure their energy needs.

This is mainly due to the fact that the development of local gas in Senegal, as wellas most of the hydro projects in Guinea (over 1300 MW), arrive only on themedium term.

The two figures below illustrate the need for natural gas to power the sub-region'sthermal power plants throughout the horizon of the study. These triple over the15-year horizon considered and pass from 1133 Mmscfd in 2017 to 3470 MmscfdIn 2033 (cfr the table listed below). However, there is a slight expected decreaseof these needs during the year of commissioning of the Mambilla plant of 3050MW in 2024.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 31/176

Fina

lver

sion

Figure 6: Evolution of natural gas needs in the WAPP region

Figure 7: Evolution of natural gas needs in the WAPP region (except Nigeria)

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 32/176

Fina

lver

sion

Sources

Nigeria

Local

BeninWAGP

(Nigeria)

BeninWAGP(Ghana

)

WAGP

Togo

Ghana

WAGP

Ghana

LocalGhana LNG

Local CIV

VicLNG

Senegal Local

Total

2017 938 0 0 3 44 70 0 77 0 0 1133

2018 1025 0 0 4 59 77 0 75 0 0 1242

2019 1260 7 0 4 68 77 0 72 0 0 1491

2020 1350 8 0 5 80 77 0 67 0 0 1593

2021 1517 9 0 6 93 77 0 62 0 0 1776

2022 1428 11 32 6 108 77 48 61 48 0 1833

2023 1377 13 30 3 126 77 61 59 67 0 1830

2024 1196 14 29 3 144 75 45 59 87 0 1667

2025 1199 17 26 1 165 68 32 58 95 119 1781

2026 1302 16 22 0 165 70 30 58 89 122 1878

2027 1474 16 19 1 164 69 32 57 80 131 2047

2028 1641 16 17 1 162 66 33 56 73 141 2213

2029 1804 16 17 3 158 60 33 56 76 160 2389

2030 2006 16 15 3 152 56 33 56 80 169 2592

2031 2223 15 15 3 152 56 59 56 91 174 2851

2032 2465 14 15 3 155 55 84 57 109 174 3140

2033 2729 17 43 3 159 55 107 57 115 183 3470

Table 1: Natural gas needs by source (units: Mmscfd)

2.3.4. Opportunities and challenges for a 100% interconnectednetworkThe region of West Africa is characterized by disparities in terms of energyresources. Indeed, some country dispose for instance of gas resources, mainly inthe eastern part of the region (Nigeria, Ghana, Côte d’Ivoire) and soon inSenegal4. Others, further north, benefit from conditions of favorable solarirradiations for the development of photovoltaic technologies (Mali, Burkina Faso,Niger). Still others have important hydroelectric potential, as is the case forGuinea, Sierra Leone and Liberia.

These differences therefore naturally call for the setting up of a large networkinterconnecting all countries in the region. This recommendation which wasalready one of the great messages of the previous master plan, is all the moretrue with the introduction of renewable energy, such as solar photovoltaic, in theenergy mix of the region.

An interconnected network will allow the transfer of this solar energy from thenorth to the south of the region during the day, and in the opposite direction duringthe evening and at night, using hydroelectric or thermal power plants.

4 Following the discovery of a gas field on the site of Grand-turtle-Ahmeyim on the border between Senegal andMauritania.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 33/176

Fina

lver

sion

On the short-term horizon considered in this section, most of the interconnectionprojects are decided, which will be addressed in detail in the transmission masterplan. At this point, it can be mentioned that these interconnections will allow tolower the marginal cost of the region from 96 USD/MWh in 2017 to 75 USD/MWhin 2022, a decrease of more than 21% over 5 years.

The distribution of the marginal costs in 2022 is shown in the figure below.

The lowest marginal costs are observed in the south-east of the region (Côted’Ivoire, Ghana, Togo, Benin and Nigeria) which has access to gas resources andhas developed numerous combined cycle projects on the 2022 horizon. Niger isalso part of the low marginal cost countries, given the development of the coal-fired power plant in Salkadamna.

Then the countries in which many renewable projects are developing (Guinea,Burkina Faso, Mali) have higher marginal costs given the use of thermal unitsusing heavy fuel that need to be activated when the renewable is no longerusable5.

Finally, countries in the western part of the region are facing the most significantmarginal costs, given the use of heavy fuel thermal units running in base(Senegal, The Gambia, Guinea Bissau, Sierra Leone, and Liberia).

Figure 8: Distribution of average marginal costs by country in 2022

5 Night for solar photovoltaic units, or dry season periods for hydroelectric units.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 34/176

Fina

lver

sion

The main challenge for the medium and long-term horizon that follow in thisanalysis will be to continue to develop the network in order to be able to exploitand share the different resources of the region : the hydroelectric power with theexpected implementation of many hydroelectric projects on the medium termhorizon; in terms of solar resources whose exploitation with photovoltaictechnology will expand massively in the medium and long term; finally, in terms ofgas resources with new resources in Senegal for which the exploitation issupposed to begin in 2025.

2.4. Optimal medium-term investment Plan 2023-2029

The medium-term horizon considered in this master plan ranges from 2023 to2029. Over this period, most of the projects that will be implemented are candidateprojects, since the main part of the decided projects was put into service before2023.

Indeed, on the 13 721 MW of decided generation projects identified during thedata collection, 8 386 MW are expected to be put in service before 2023, whichleaves 5357Mw for the medium and long-term horizon.

On these 5357 MW of decided projects, the most important is the hydroelectricplant of Mambilla In Nigeria, with an installed capacity of 3050 MW. Other projectsinclude 1322 MW of hydroelectric projects of smaller size, 2x350 MW for the twophases of San Pedro coal plant in Côte d’Ivoire and 285 MW for the ALAOJI 2+gas plant in Nigeria.

Figure 9: Distribution of the projects decided in the medium term for WAPP per fuel type

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 35/176

Fina

lver

sion

Given that the peak load is supposed to increase from 21 331 MW in 2022 to 36397 MW in 2029 according to the load forecast presented in tome 2, the decidedprojects will not guarantee the security of supply in the region. Therefore, anadditional investment of 9868 MW in conventional units is required (8872 MWthermal and 996 MW hydro).

Besides these investments in conventional plants, the master plan has identifiedup to 15 828 MW of potential solar projects that should allow to reduce the costof electrical energy drastically and contribute to the development of sustainabledevelopment in the region.

Of these 15.8 GW of potential projects, the network studies carried out in thetransmission master plan below validated the integration of 5.4 GW, and moreprecisely, 3.4 GW between 2022 and 2025 and 2.0 GW between 2025 and 2029,thus bringing the installed capacity in solar photovoltaic at 6.8 GW in 2025 and8.8 GW in 2029. Further technical studies will however be necessary to confirmthe integration of the remaining 10.4 GW.

Finally, several potential wind projects have been also identified at the end of themedium-term horizon considered here.

Figure 10: Distribution of the projects decided in the medium term for WAPP, per fuel type, including the potential solarprojects identified

The impact of this investment plan on the energy mix at the end of the mid-termhorizon (in 2029) is shown in the figure below. Solar photovoltaic technology hasthe largest increase compared to the 2022 situation, accounting for 13% of theregion's total energy generation if the total identified potential is actually exploited(corresponding to an installed power of 19.2 GW in 2029). However, taking intoaccount only solar projects whose integration has been technically proven(corresponding to an installed power of 8.8 GW in 2029), the contribution of thistechnology to the mix of the sub-region would be 6%.

Gas-fired power plants still produce most of the electricity (60%). Hydroelectricityis stable and revolves around 24% generation. Wind technology appears in themedium-term energy mix, but still weakly with only 1% of the total energyproduced in the region.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 36/176

Fina

lver

sion

The other highlight is the almost disappearance of the use of heavy fuels anddiesel on the 2029 horizon. The exploitation of gas resources now both in the eastand in the west of the region6, as well as the development of interconnectionsbetween the different member states allow to do without these expensive andpolluting fuels in the medium term.


Figure 11: Energy Mix of the WAPP at then of the medium-term, including the potential solar projects identified

2.4.1. The exploitation of regional hydropower potential: a priorityThe mid-term period referred to in this section is characterized by thecommissioning of many hydroelectric power plants. Indeed, as was mentioned inthe introduction, over 5000 MW has to be put into service on the 2023-2029horizon (of which a majority of decided projects).

The countries concerned with this development are mainly Nigeria, Guinea andSierra Leone. Some projects are also planned in Côte d'Ivoire and Togo. The listof these projects can be found in the table below.

6 Following the discovery of a major gas field on the border between Senegal and Mauritania

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 37/176

Fina

lver

sion

Country Project Status TechnologyInstalledpower(MW)

Commissioning

Côted’Ivoire Louga Decided Hydro 224 2023

Côted’Ivoire TIBOTO Decided Hydro 112.5 2026

Guinea Amariah Decided Hydro 300 2023

Guinea MORISANAKO Selected Hydro 100 2023

Guinea GRAND KINKON7 Selected Hydro 291 2023

Guinea KOUKOUTAMBA Decided Hydro 294 2024

Guinea BONKON Diaria Selected Hydro 174 2025

Guinea TIOPO Selected Hydro 120 2028

Guinea DIARAGUÔLA Selected Hydro 72 2029

Nigeria Mabon Selected Hydro 39 2023

Nigeria MAMBILLA Decided Hydro 3050 2024

SierraLeone BUMBUNA II Decided Hydro 132 2023

SierraLeone

BUMBUNA III(Yiben) Decided Hydro 66 2023

SierraLeone BENKONGOR I Selected Hydro 34.8 2023

SierraLeone BENKONGOR II Selected Hydro 80 2025

SierraLeone BENKONGOR III Selected Hydro 85.5 2026

Togo ADJARALA Decided Hydro 147 2026

Togo SARAKAWA Decided Hydro 24.2 2023

Total Hydro 5357

Table 1: List of hydroelectric projects over the medium-term

Among all these projects, Mambilla in the east of Nigeria is the one with the mostimportant size, with 3050 MW. In Guinea, the master plan foresees thecommissioning of 1351 MW in the medium term including Amaria (300 MW),Koukoutamba (294 MW) and Grand Kinkon (291 MW). In Sierra Leone, the sitesof Bumbuna and Benkongor will welcome 400 MW. Finally, we will also note thedecided projects of Louga and Tiboto in Côte d’Ivoire and Adjarala in Togo.

7 Recent studies have determined that the capacity of the Grand Kinkon unit could be somewhat lowered

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 38/176

Fina

lver

sion

2.4.2. Strong integration of renewable energies for an optimalenergy mixAs we have shown above, the share of renewable in the energy mix over themedium-term horizon will grow significantly. In fact, in 2029, the electricityattributable to renewable energy sources could reach up to 38%, including 24%hydropower, 13% solar photovoltaic and 1% wind energy.

Risk analysis for the integration of intermittent renewable energies

In the absence of any constraint, economic optimization selects the option at lowercost to meet the demand on the horizon of the study. In this context, there is acost threshold below which the PV solar option becomes marginally moreinteresting than the thermal options in West Africa. The existence of this level hastwo effects:

§ On the one hand, before the crossing of this threshold which will intervenearound 2025 according to the assumptions of the master plan, investments inPV solar projects in the subregion remain marginal if one refers to economiccriteria only

§ On the other hand, at the pivotal year (year when the investment threshold iscrossed), the investment volumes are considerable (several tens of GWinvested in a single year)

Taking the risk into account in the master plan allows us to deviate from thisscenario, which, although economically optimal, is difficult to implement inpractice. In order to take into account the limited financial capacities of theMember States of the WAPP, particularly in view of the fact that renewableprojects have a high capital intensity; in order to take into account also thetechnical limits linked to the integration into the still fragile network of the WAPP,it was decided to set a limit on the annual investment by country. As such, themacimum capacity that can be invested each year for each technology (wind andsolar PV) has been defined at 10% of national peak demand. This limit has severaleffects:

§ It has been defined in such a way as to achieve, on the horizon of the study,a volume of renewable investment close to that observed in the optimal casefrom the economic point of view

§ It allows a better distribution of investments in time, before and after the pivotalyear

§ It allows for a better geographical distribution of investments, as the countriesof the north of the region are no longer solely responsible for all renewableprojects

The optimization program used to carry out this master plan has identified animportant need for potential solar photovoltaic projects over the medium-term upto 15.8 GW over the 7 years considered, which represents more than 2 GW eachyear in the region.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 39/176

Fina

lver

sion

The geographical repartition of the investments depends, on one hand, of thesolar irradiance of the country, and on the other side of the level of the demand,the latter point having the objective to spread the investments in order to limit therisks and to facilitate the access to capital. As such, important potential solarprojects have been identified in Burkina Faso, Mali and Niger, but also in thenorthern region of Nigeria, Benin, Togo, Ghana, and Côte d’Ivoire. Moreover, theexpected fall in costs of the PV technology implies that investments in the southof the region are beginning to be profitable on the end of the mid-term horizon(i.e., from 2026).

Concerning wind turbine technology, some projects are starting to be justified atthe end of the medium-term horizon, notably in Nigeria

The typical economic dispatch of the peak day in 2029 is shown on the figurebelow, taking into account the whole potential solar projects identified. Thepredominance of solar photovoltaic energy appears clearly during the day, thelatter providing almost 50% of the energy consumed in the region at midday.

This figure illustrates also the possible synergy between hydroelectric and solarresources. Hydroelectricity is mainly reserved for the evening peak as well asduring the night, when solar energy is unavailable. During the day, though, somehydroelectricity is produced, given the technical minimum imposed, especially forirrigation issues.

Figure 12: Typical dispatch at the end of the medium-term (2029), including the potential solar pojects identified

2.4.3. Diversifying thermal resources to limit exposure to risk andvolatilityAlongside the investments in renewable, it is essential to continue to invest inthermal units in order to meet the ever-growing demand of the region whenrenewable resources are unavailable, but also to increase the reliability of thepower system.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 40/176

Fina

lver

sion

The medium term is thus also characterized by the development of gas-firedpower plant, and this, not only in the eastern part of the region, but also in thewestern part, following the discovery of a gas field in Senegal.

The region's needs in thermal units on the medium-term horizon are estimated to9.2 GW. The bulk of this capacity is to be developed in Nigeria, with nearly 8 GWto install between 2023 and 2029. For Senegal, the optimization proposes thedevelopment of combined cycles for a total power of 750 MW in 2025. Finally, acombined gas cycle project of 450 MW is also recommended in Ghana8 for theend of the period (i.e. on the 2029 Horizon).

In Nigeria, 4.8 GW of combined cycles have been identified during the datacollection (including Egbin, Ethiopia, Caleb Inland, Alaoji, Geregu, Omotosho,Calabar Odukpani and Gbarain Ubie). Nevertheless, this is insufficient to coverthe 8 GW needed over the period. Therefore 3.2 GW of complementary combinedcycle projects are necessary in Nigeria between 2023 and 2029. From a practicalpoint of view, it is recommended to develop these projects on sites of existingopen-cycle gas plants in order to transform them into combined cycle.

As far as coal is concerned, we will note the commissioning in 2026 of the coal-fired plant of San Pedro I in Côte d’Ivoire, with a capacity of 350 MW, with asecond phase foreseen in 2029.

The geographical diversity of the proposed thermal projects and the variety offossil resources used (domestic or imported) allow for a greater security of supply,guaranteeing the electricity supply of the sub-region even in case of lack of supplyfrom one of the sources;

2.4.4. The interconnected network to better share the resourcesThe potential renewable energy projects in the medium term as well as thedevelopment of the interconnected network pulls the marginal costs of the wholeregion downwards. These could indeed evolve from 80.6 USD/MWh in 2022down to 49 USD/MWh in 2029.

8 Potentially on the site of Aboadze

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 41/176

Fina

lver

sion

Figure 13: Evolution of average marginal costs by country between 2022 and 2030

Beyond this widespread decline in marginal costs in the region, it can be notedalso that the latter vary greatly depending on the time of the day considered.Indeed, if we analyze the situation during the day at 12h, the marginal costs asshown in the figure below are observed. These are naturally lower in the north ofthe region (where the electricity comes mainly from solar photovoltaic technology)as well as in Guinea and Sierra Leone in which many hydroelectric power plantsare now active.

We therefore clearly see the importance of developing the interconnected networkso that we can share these renewable resources and in particular the axesattached to Guinea, as well as the southern backbone connecting Nigeria toBenin, Togo, Ghana and Côte d'Ivoire and the north backbone between Nigeria,Niger and Burkina Faso.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 42/176

Fina

lver

sion

Figure 14: Distribution of the average marginal costs at 12h in 2025

During the evening peak of 9 pm, the situation is different. The northern countries,such as Mali, Burkina Faso and Niger must now import, given the lack of solarenergy. They therefore face the most important marginal costs at this moment.These imports come mainly from the countries disposing of gas resources whichuse their combined-cycle power plants (mainly Nigeria, Ghana, Côte d’Ivoire andSenegal). We observe on the figure below that the flows are effectively reversedin the evening.

Thus, Nigeria exports during evening via the south and north backbones, whileSenegal exports to Mali and The Gambia. Let's note that the situation aroundGuinea remains relatively stable throughout the day (i.e. a situation of quasipermanent export to its neighbouring countries).

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 43/176

Fina

lver

sion

Figure 15: Distribution of the region's average marginal costs at 21h in 2025

These considerations regarding the exchanges will be discussed in more detailsin the the transmission master plan below, but it already shows the major trendsthat will guide the evolution of the network.

2.5. Optimal long-term investment plan 2030-2033

The long-term period considered in this study covers the years 2030 to 2033. Overthis interval, demand continues to grow exponentially in the region. The forecastof the synchronous peak demand of the region evolving from 36.4 GW in 2029 to50.8 GW in 2033.

The investments needed to cope with this surge in demand are estimated to14981 MW for thermal generation and 562 MW for hydroelectric generation.

Besides those conventional projects, potential solar PV and wind projects havebeen identified in the context of this master plan, for an amount up to 16 700 MWfor solar PV and 750 MW for wind turbine.

Concerning solar projects, the technical simulations carried out in thetransmission master plan developed in Chapter 3 below have validated theintegration of 1.9 GW between 2030 and 2033, bringing the total solar capacity ofthe subregion to at least 10.7 GW at the end of the study. In-depth technicalstudies will however be necessary to confirm the integration of the remaining 14.8GW over 2030-2033.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 44/176

Fina

lver

sion

Figure 16: Distribution of long-term investments by fuel type, including the potential renewable projects identified

Compared to the mid-term period, it can be seen that investments in hydroelectricplants are proportionately very small. Indeed, the most economically interestingprojects have been taken into account in the mid-term and the remaining projectsdo not appear to be an economically viable option for the sub-region's powersupply up to the 2033 horizon.

On the other hand, the investments in thermal power plants are proportionatelymore important. This is required to guarantee the reliability of the interconnectednetwork, these challenges of reliability increasing as the renewable share takes agrowing importance in the energy mix of the region.

The energy mix at the end of the study horizon is illustrated below. The share ofgeneration attributable to renewable energies is 36% in 2033, of which 18% forhydroelectric power plants, 17% for solar photovoltaic, if the entire solar potentialprojects were effectively developed (corresponding to an installed capacity of 36GW in 2033) and 1% for wind energy.

However, if one takes into account only solar projects whose integration has beentechnically proven (corresponding to an installed power totalling 10.7 GW in2033), the contribution of this technology to the mix of the subregion would gofrom 17% to 5 percent.

Still, most of the electricity generated in the region comes from the gas-fired plants(62%), of which 77% is provided by Nigeria. Coal-fired power plants finallyproduce 2% of the total energy.


WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 45/176

Fina

lver

sion

Figure 17: Energy mix of the region at the end of the study (2033) , including the potential renewable projects identified

2.5.1. Towards optimal exploitation of economically profitablehydroelectric resourcesSome hydroelectric projects come out of the optimization on this long-termhorizon, including 3 in Guinea (Boureya, Fetor and Lafou) and 1 at the borderbetween Liberia and the Sierra Leone (Mano). These plants amount to 562 MWin total, to which one can add the Saint-Paul river projects for an expectedcapacity between 360 and 585 MW, which is currently under study.

Country Project Status Technology FuelInstalledpower(MW)

Commissioning

Guinea Boureya Selected Hydro Hydro 160 2030

Guinea Fetor Selected Hydro Hydro 124 2031

Guinea Lafou Selected Hydro Hydro 98 2032

Liberia Mano Selected Hydro Hydro 180 2032

Liberia Saint-Paul Under study Hydro Hydro 360-585 >2030

Table 2: Investments in hydro projects in the long term

These projects are selected in the current master plan in the light of theireconomic interest, but also for their ability to compensate for the variability ofrenewable energies (solar and wind turbines).

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 46/176

Fina

lver

sion

Nevertheless, it is necessary to mention that the two reasons given above (i.e.economic reason and integration of renewable) are not the only ones justifyinginvestments in hydroelectric power plants. Therefore, some projects that don'tcome out of the present optimization, and which therefore do not have a regionalvocation, could nevertheless be developed for other uses (such as irrigation forinstance).

2.5.2. Towards a meshed networkThe potential solar photovoltaic projects identified in northern regions of theWAPP as well as the development of the hydro potential in the west of the region(Mainly around Guinea) make even more obvious the differences in marginalcosts in the middle of the day at 12h, as can be seen in the figure below.

Figure 18: Average marginal costs at 12h at the end of the study (2033)

It is actually observed that the west and north of the region are characterized bythe lowest marginal costs at noon given the abundance of solar and hydroresources.

In the west of the region, hydroelectric resources are used, mainly in Guinea,Sierra Leone, Liberia and the Mali. It makes sense that these resources areshared via the OMVS, OMVG and CLSG networks.

The east of the region is exposed to higher marginal costs, particularly in Nigeriawhere gas-fired power plants are running in base. Therefore, there is an economicinterest in exporting this low-cost electricity generated in the countries from thenorth (Burkina Faso and Niger) and west (Côte d’Ivoire and Ghana) to Nigeria,through Togo and Benin.

This reinforces the recommendation done in the medium term to develop thenetwork in this eastern part of the region, to transfer power from the northern andwestern countries to Nigeria (through Togo and Benin).

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 47/176

Fina

lver

sion

At the evening peak (at 9pm), the situation is virtually reversed (see figure below).Indeed, we see that the countries in the north which used mainly solar energy atnoon (such as Mali, Burkina Faso and Niger) are now facing far greater marginalcosts. These countries will now import mainly from the countries which have gasresources (Nigeria, Ghana, Côte d'Ivoire and Senegal) or Hydropower (Guineamainly).

Figure 19: Average marginal costs at 9pm at the end of the study (2033)

2.5.3. Flexibility and reliability issues on the long-termThe present master plan is designed to bring the Loss of Load Expectation(LOLE9) to 24 hours a year in the whole region at the end of the study in 2033,while considering an interconnected system where mutual support is possible tocompensate for the lack of generation in a given country.

The LOLE is a probabilistic criterion that indicates the expected number of hoursin a year in which the demand exceeds the available generation capacity, resultingin the inability to provide the full load without mitigation measures.

This criterion has been integrated into the simulations. The result is that additionalinvestments must be made in thermal units essentially from 2030. This explainsin particular why the investments in gas power plants are proportionately moreimportant on the long-term horizon compared to other horizons, as was mentionedat the beginning of this section. These investments are included in appendix.

9 Loss of load expectation

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 48/176

Fina

lver

sion

Regarding the combined cycles, Nigeria accounts for 9000 MW. A 369 MW CCproject is also recommended in Côte d’Ivoire (Songon). Lastly, it is recommendedthat Senegal continue its development of combined-cycle power plants up to 900MW in the long run.

Alongside these combined cycle projects, we see that the generation master planalso provides the development of open-cycle gas turbines, particularly in Nigeriafor 3500 MW, in Ghana for 300 MW and Côte d’Ivoire for 300 MW. Theseinvestments are necessary for flexibility reasons, but also for reliability reasons,in order to guarantee compliance with the LOLE of 24h/year.

Also, it should also be mentioned that, for the same reasons of flexibility andreliability, but also for reasons of security of supply, it is recommended to thecountry which have no gas resources to develop combined cycles of small size(typically 60 MW). These are included in the detailed table of the geenrationmaster plan in Appendix A of this document.

All these extra investments in thermal units imply that the marginal costs remainstable from 2030 to 2033, around 49 USD/MWh.

Finally, the battery storage will have to play a major role in improving flexibilityand increasing the security of supply of the sub-region. In view of the expectedchanges in the cost of batteries, they could supplant part of the investments ingas turbines on the horizon of the study.

2.6. Synthesis

In order to meet the electrical demand, supposed to reach 50.8 GW in 2033, thepresent master plan indicates that the installed capacity of thermal units shouldreach 45.4 GW in 2033 and the hydroelectric plants 12.8 GW.

In order to reduce the costs of electricity and to reduce the ecological footprint ofthe sector, several renewable energy projects (solar PV and wind turbine) havebeen identified and recommended for a total capacity reaching 37.5 GW. Thesimulations carried out in the transmission master plan in chapter 3 belowvalidated the technical integration of 12.1 GW of these intermittent renewableprojects at the end of the study. The development of the additional 25.4 GW (tocover the 37.5 GW) can be carried by the countries and will therefore requireadditional technical studies.

Regarding the energy mix, the solar photovoltaïc technology would contribute upto 17% of the energy generated in the region at the horizon 2033, if the entirepotential solar projects were to be developped.

The large part of gas in the energy mix is also striking, which replaces HFO andDDO at the horizon of the study. Finally, it appears that the part of hydroelectricityin the mix, after exhibiting an increase over the medium term, finally tends todecrease, given the fact that all the economically interesting projects wereimplemented.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 49/176

Fina

lver

sion

Figure 20: Evolution of the energy mix (in GWh), taking into acocunt the entire solar projects identified

Figure 21: Evolution of the energy mix (in %)), taking into acocunt the entire solar projects identified

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 50/176

Fina

lver

sion

3. TRANSMISSION MASTER PLAN

The aim of this chapter is to present the transmission network and its evolutionover the study period. This analysis directly follows the economic analysis and theobjective is to validate that the economic results are technically feasible over thestudy period.

This technical feasibility will be studied over 3 different time horizons (selectedwith the Member States during the project meeting in May 2018); respectively2022, 2025 and 2033. Different scenarios will be evaluated over these differenttarget years in order to test the system and asses its limits. Static analysis will becarried out for all three target years and dynamic analysis will be carried out forthe two first target years.

The 2022 time horizon which comprises mainly of decided generation andtransmission projects will be studied. The main objective of this study year is toanalyze and identify the weakest points of the network in order to connect thedifferent blocks together which are currently not synchronized. The conclusionsof these analysis will serve as the basis for the priority investments which shouldbe conducted in order to synchronize the region securely.

The objective of the study year of 2025 is to verify that the economic exchangesplanned are feasible on the technical point of view and to solve the weak pointsdetermined in the study of the year 2022 in order to create a network which canoperate within the operational limits under the N-1 condition.

Finally, the year 2033 will be studied to determine the reinforcement needsexpected in the long term which would satisfy the economic exchanges and thelevel of renewable integration given from the economic analysis. The referencegrid structure of 2033 has been built on:

· For the national power system: information extracted from the master planstudies and collected during the data collection phase;

· For the reinforcement of the interconnections, the economic study assessedthe optimal size of transfer capacities between the different areas.

This chapter first described the methodology adopted and the assumptions donefor the modelling of the WAPP network. The different scenarios and the resultsfrom the different analysis are then explained in the rest of the chapter and thedifferent investment priorities are detailed.

3.1. Methodology

3.1.1. Static analysisThe objective of the static analysis is to visualize the flows on the lines andvoltages at all substations in the interconnected network. Based on the load flowresults, the necessary reinforcement can be performed in order to best satisfy theoperational constraints.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 51/176

Fina

lver

sion

The N-1 security criteria is applied starting from the study year of 2025. Theoperational limits allowed in normal situation (N) and in single contingencysituation (N-1) is shown in the following table. The list of contingencies analyzedconcern all equipment of voltage levels 225 kV and higher.

Maximum load of transmission infrastructure

Works State N

(Normal Situation)

State N-1

(Under Incident)

% nominal power % nominal power

Lines 100 % 110 %

Transformers 100 % 120 %

Voltages ± 5 % ± 10 %

Table 3: Operational limits

3.1.1.1. MODEL CREATION

The model of the high voltage grid was built starting from the 2017 existingsituation which was validated with each country during the dedicated workshop.

The objective of the WAPP Master Plan is to identify, on the regional level, thereinforcement needs and to challenge the existing high voltage grids of eachcountries against the interaction that could appear between countries. For thisreason, only the parts of the network which are meshed and which could bedirectly impacted by the interconnection of the West African countries has beenmodelled.

The Table 4 below summarizes the voltage level that were modelled for eachcountry. The picture below shows a typical example of the modelling that wasdone. The high voltage (red in the example) network was modelled as well as themeshed network level (green in the example) which can be impacted in theregional exchange of power. The load is connected on the low voltage levelthrough distribution transformers.

The full load flow model as it was developed is shown in Figure 23 where the toppart represents the Western part of the WAPP and the bottom part represents theEastern part. It is clear that these two networks are interconnected under onesingle synchronous model in the studies that were performed.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 52/176

Fina

lver

sion

Figure 22: Typical detail level of modelling of a country’s high voltage grid

Country Voltage Levels Modelled

Senegal 225 & 90 kV

The Gambia 225 & 132 kV

Guinea Bissau 225 kV

Guinea 225 & 110 kV

Mali 225 & 150 kV

Liberia 225 kV

Sierra Leone 225 & 161 kV

Côte d’Ivoire 225 & 90 kV

Ghana 330 & 161 kV

Togo 330 & 161 kV

Benin 330 & 161 kV

Burkina Faso 225 & 90 kV

Niger 330 & 132 kV

Nigeria 330 &132 kV

Table 4: Voltage levels modelled for each country

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 53/176

Fina

lver

sion

Figure 23: Full load flow model of the WAPP (top=West, bottom=East)

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 54/176

Fina

lver

sion

3.1.1.2. LOAD

The load level which is modelled is based on the results of the load demandforecast presented in Tome 2.

The power factor which was modelled in the model of the existing network (2017)was used for the following years and kept constant until 2025. Load with a powerfactor of less than 0.9 were however set to 0.9 in 2033. For this study year of2033, the power factor of Nigeria was adapted to 0.95 accordingly to the TCNMaster Plan.

For each country, national master plans were used when available in order toevaluate the spreading of the future loads on the high voltage grid.

3.1.1.3. GENERATION

Each generation unit is modelled behind its step-up transformer as shown in thefollowing Figure 24.

Figure 24: Connection model of generators

The modeling of new power plants was done considering that rotating machines(Hydro units, thermal units, and biomass) have a cosφ of 0.85 for their productionof reactive power and a cosφ (power factor) of 0.95 in absorption. The step-uptransformer is modelled with an apparent power of the size of the unit that isconnected.

Renewable generation such as wind turbines and PV parks are modelled with acosφ of 0.95 in both absorption and production of reactive power. It is assumedthat new technology to this day allows such reactive power range and newly builtcapacities should be installed with the specificities.

3.1.1.4. TRANSMISSION

New transmission lines are modelled using the same parameters (per km) assimilar lines recently constructed in the country or region of interest.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 55/176

Fina

lver

sion

3.1.2. Dynamic AnalysisThe objective of the dynamic analyses is to assess the stability margins of thesystem to ensure its secure synchronous operation. The results of the staticanalysis is taken as input. The analyses carried out are the following:

· Small Signal Stability analysis: aimed at evaluating the response of the systemto small variations typical of normal operation, such as breaker switch and loadvariation. The outcome of the analysis will highlight power oscillationsinsufficiently damped that might endanger the stability of the system.Methodology is detailed in Annex.

· Dynamic Security analysis: aimed at evaluating the ability of the system toendure electrical transients and recover a sustainable operating point. Thisanalysis can be seen as an extension of the N-1 security criteria. Following theloss of several key transmission network elements, the transient and voltagestability of the system will be assessed. Methodology is detailed in Annex.

· Frequency Stability analysis: aimed at evaluating the ability of the system toendure transients involving unbalances of power and to stabilize systemfrequency without activating defensive measures. In this analysis, the inertialand primary response of the system is tested. Methodology is detailed inAnnex.

A description of the dynamic model implemented is presented in Annex.

3.2. Short term development plan - 2022

3.2.1. Towards an interconnected systemThe West African electrical network being currently operated in multiple differentblocks, exchanges between countries are currently limited to the sharing of theresources between one country and its close neighbors. This sharing of resourcesrequires high voltage transmission lines to be built between countries in orderto allow for the sharing of resources from the most Eastern part of the region tothe most Western part (from Nigeria to Senegal) and the operation of the WestAfrican power grid in one single synchronous network.

Currently, as of January 2018, the different synchronous blocks are as follows:

· Block A: Burkina Faso, Ghana, Côte d’Ivoire, part of Mali (up to Bamako) andpart of Togo/Benin.

· Block B: Senegal, Mauritania and part of Mali (up to Bamako)· Block C: Nigeria, Niger and part of Togo/Benin· The other countries of the WAPP are not connected to each other through the

High Voltage (HV) grid and operate in an isolated way. These countries areGuinea, Guinea Bissau, The Gambia, Liberia and Sierra Leone.

In the short term, these three asynchronous blocks are due to be interconnectedin order to function under one synchronous interconnected system. Additionally,countries currently isolated will be connected to the single synchronous zone,allowing them to share with and profit from their neighboring countries’ resources.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 56/176

Fina

lver

sion

Currently, the three existing blocks which are operated asynchronously are beingoperated in this matter due to technical and stability issues. In fact,interconnection lines are constructed and available, but the currently unstablenature of the system does not make it possible to operate under one synchronousarea.

In the short term, the objective of operating under one synchronous network ismainly related to the stability of the system. In order to increase the systemstability and reach a safe operation of one synchronous zone, the commissioningof new interconnection lines is primordial. Secondly, the importance of the ICC ishere put forward due to its different roles in order to operate an interconnectednetwork. These two aspects are clearly a necessity in the short-term horizon andshould be a priority to operate the WAPP network synchronously.

In terms of stability, the synchronous operation of the WAPP system will be themost challenging in its first years of operation. The network presents longdistances, different topologies and few weakly interconnected borders, generallyat the interface between the current synchronous blocks.

The major stability issues have been detected at the interfaces between theexisting synchronous blocks, in particular:

· The interconnected WAPP system will be subject to interarea modes (largegroups of generation units swinging in opposition to each other following smalldisturbances on the network) due to long distances. The recently finalized“WAPP synchronization study” already recommended installing additional PSSbut this measure should be extended before the synchronisation of the entireWAPP system with dedicated tuning to damp interarea oscillations. Inparticular, a dangerous interarea mode has been detected between thesynchronous block C and the rest of WAPP.

· The interface between Block C and B is not secure. The stability of the entiresystem is compromised whenever one of the lines is faulted. A SpecialProtection Scheme (SPS) is recommended in order to operate safelyconsidering the economic exchanges planned between Nigeria and the rest ofthe WAPP.

· The interface between Block A and B is not secure as it’s composed of onlytwo single-circuit transmission lines. The stability of the entire system iscompromised whenever one of the lines is faulted. A minimum set ofreinforcements is proposed to secure the interface.

· Voltage support is insufficient in different zones of the WAPP system. Oncesynchronized, voltage instabilities can easily evolve in system-wide issues.Thus, the most problematic zones have been identified and remedial actionsproposed.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 57/176

Fina

lver

sion

The criticalities identified in the static and dynamic studies are represented in thefigure below.

Figure 25: Transmission network criticalities at short-term - 2022

3.2.2. Modelling of the 2022 WAPP networkThe hypothesis and information that was used to create the 2022 model of theWAPP are described in the following sections.

3.2.2.1. DECIDED INTERCONNECTION PROJECTS

The model of the existing grid was taken as the starting point to create the 2022model of the WAPP network.

In the short-term horizon of 2022, the following interconnections shown in Table5 have been decided and will be present. These interconnecting lines areconsidered to be constructed and fully operational by 2022. In such a way, blockA will be connected to block C with the commissioning of the Sakete (Benin) –Davié (Togo) – Volta (Ghana) 330 kV line and the North Core (Nigeria – Niger –Burkina). The commissioning of the OMVG line and the Guinea – Mali line willinterconnect Guinea to Mali and to block A and B. Additionally, thisinterconnection between Guinea and Mali will create a direct link to CLSG. Theisolated countries of Sierra Leone and Liberia will be connected through the firstcircuit of the CLSG line. The same saying is true for the countries of The Gambiaand Guinea Bissau which will be interconnected through the OMVG line.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 58/176

Fina

lver

sion

Country HV Interconnection VoltageLevel[kV]

RatedPower[MVA]

CommissioningYear

Length[km]

Single (SC)/ Double

Circuit (DC)

CI-LI-SL-GU CLSG 225 330 2020 1303 DC

GH-BU Bolgatanga-Ouagadougou

225 330 2018 198 SC

GH-TO Volta-Davié (Lomé) 330 1000 2019340

SC

TO-BN Davié-Sakete 330 1000 2019 SC

BU-NR-NI-BN Dorsale North 330 777 2022 832 DC

SE-GA-GB-GU OMVG 225 330 2020 1677 SC

MA-SE Kayes-Tambacounda 225 330 2020 288 DC

SE-MAU Noukchott-Tobene 225 330 2020 425 DC

MA-MAU Kayes-Kiffa 225 330 2021 420 DC

TO-BN Porga-Dapaong 161 178 2022 83 SC

TO-GH Dapaong – Bawku 161 178 2022 53 SC

GU-MA N’Zérékoré – Fomi –Bamako

225 330 2022 1074 DC

Table 5: Decided interconnections (After 2017)

3.2.2.2. NATIONAL REINFORCEMENTS

Additionally to these interconnections, the elements which are added to theexisting model to create the 2022 network are summarized by country in theAnnex. These reinforcements are based on the national and regional masterplans and based on the information collected during the model validationworkshop. It should be noted that most of these national projects were consideredas decided and were not subject to an optimization due to the short-term horizonof 2022 and the regional vision of this master plan.

3.2.2.3. DEFINITION OF THE SCENARIOS – TARGET YEAR 2022

The load levels which were modelled in the different scenarios that were studiedfor 2022 are details in the following paragraph.

For the target year of 2022, two different scenarios were analyzed:

· Asynchronous peak evening situation· Synchronous Off-peak with maximum renewable infeed scenario

In the peak load scenario, the load modelled corresponds to the yearlyasynchronous peak load of every country. This scenario where every countryobserves their peak load at the same time is a conservative way of analyzing thereinforcement needs on the grid. This active load level is presented in Table 6.The same power factor as the existing model (2017) was kept for this study year.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 59/176

Fina

lver

sion

Country Peak Load 2022

BENIN 359 MW

BURKINA 471 MW

CÔTE D’IVOIRE 2013 MW

GAMBIE 140 MW

GHANA 3217 MW

GUINEE 551 MW

GUINEE BISSAU 105 MW

LIBERIA 166 MW

MALI 680 MW

NIGER 430 MW

NIGERIA 11500 MW

SENEGAL 944 MW

SIERRA LEONE 428 MW

TOGO 328 MW

TOTAL 21331 MW

Table 6: Load level - Peak 2022

The load level modelled in the off-peak scenario is such as shown in Table 7. Thisload level is the lowest yearly synchronous load.

Country Off-peak Load 2022 Percentage ofpeak load (%)

BENIN 165 MW 46 %

BURKINA 269 MW 57 %

CIV 1047 MW 52 %

GAMBIE 62 MW 44 %

GHANA 2155 MW 67 %

GUINEE 242 MW 44 %

GUINEE BISSAU 46 MW 44 %

LIBERIA 73 MW 44 %

MALI 415 MW 61 %

NIGER 228 MW 53 %

NIGERIA 5980 MW 52 %

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 60/176

Fina

lver

sion

Country Off-peak Load 2022 Percentage ofpeak load (%)

SENEGAL 548 MW 58 %

SIERRA LEONE 188 MW 44 %

TOGO 151 MW 46 %

TOTAL 11568 MW 51 %

Table 7: Load level - Off-peak 2022

3.2.2.4. NORTH CORE COMPENSATION

The compensation scheme of the North Core project has been reviewed recently.In addition to the shunt compensation already foreseen and included in the model,serie compensation could be added on the line. This series compensation is notincluded in the study model. However it should be noted that it could impact themaximum power transfer between Niger/Nigeria and WAPP and the requiredmeans in terms of voltage control, albeit not to such extent as to change theresults of this regional planning study.

3.2.3. Static studiesThis section presents the simulation results for the different scenarios that weretested in 2022. The balance of each country is presented as well as the generationdispatch and the results for each scenario. A special attention will be given hereto the analysis of the synchronization of the WAPP in 2022 and the investmentsneeded in order to successfully operate the synchronous network in a stable way.

3.2.3.1. ASYNCHRONOUS PEAK 2022

In this scenario, it is considered that no renewable power is produced by PV plantsdue to the fact that the peak appears during the evening hours. This aspect isrepresented in Figure 26.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 61/176

Fina

lver

sion

Figure 26: Load vs solar irradiation curve

The available generation units as well as the peak dispatch are based on theresults of the economic analysis. The country balances which are represented inTable 8 are a direct consequence of the results of the economic analysis. In thispeak scenario, similarly to solar power, wind power is also assumed to beinexistent. Furthermore, it is assumed that hydro power plants are dispatched attheir maximum capacity.

Country Peak Balance

BURKINA - 361 MW

CIV 141 MW

GAMBIE - 28 MW

GHANA 21 MW

GUINEE 464 MW

GUINEE BISSAU - 62 MW

LIBERIA - 84 MW

MALI - 60 MW

NIGER - 62 MW

NIGERIA 663 MW

SENEGAL - 91 MW

SIERRA LEONE - 299 MW

TOGO - BENIN - 229 MW

Table 8: Country balance – Asynchronous Peak 2022

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 62/176

Fina

lver

sion

Line Name Voltagelevel (kV)

Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)

Ikeja West_330-Sakete_330-1 330 NI TB 302.6 52.9

Goroubanda_330-Ouaga Est_330-1 330 NR BU 140.7 44.4


Birnin Kebbi_330-Zabori_330-1 330 NI NR 129 40.7

Birnin Kebbi_330-Zabori_330-2 330 NI NR 129 40.7

Linsan_225-Kamakwie_225-1 225 GU SL 117.4 34.5


Kayes_225-Bakel_225-1 225 MA SE 95.4 37.4

Katsina_132-Gazaou_132-1 132 NI NR 78.5 85.2

Boke_225-Salthinho_225-1 225 GU GB 73.2 22.2

Siguiri_225-Sanakoroba_225-1 225 GU MA 66.6 20.7


Man_225-Yekepa_225-1 225 CI LI 56.4 18.5

Man_225-Yekepa_225-2 225 CI LI 56.4 18.5

Bolgatanga_330-Bobo_330-1 330 GH BU 53.5 11.3


Davié_330-Dawa_330-1 330 TB GH 49.2 5.6

Tanaf_225-Soma_225-1 225 SE GA 30.3 9.1

Mano_225-Kenema_225-1 225 LI SL 30.1 8.8

Mano_225-Kenema_225-2 225 LI SL 30.1 8.8

Birnin Kebbi_132-Dosso_132-1 132 NI NR 28.1 30.8

Cinkassé_161-Bawku_161-1 161 TB GH 27.6 16.5

Mansoa_225-Tanaf_225-1 225 GB SE 27.5 11.3

Zabori_330-Malanville_330-1 330 NR TB 24.1 7.4

Lomé (Aflao) 1_161-Aflao Ghana_161-1 161 TB GH 19.9 29.3

Bobo_330-Sikasso_330-1 330 BU MA 17.8 2.5


N'Zérékore_225-Yekepa_225-1 225 GU LI 13.9 4.3


Kayes_225-Tambacounda_225-1 225 MA SE 11.7 6.5


Ferkéssédougou_225-Sikasso_225-1 225 CI MA 11.2 3.4

Ferkéssédougou_225-Kodeni_225-1 225 CI BU 10.9 3.5

Asiekpe PST_161-Lomé (Aflao) 1_161-1 161 GH TB 3.1 17.7

Soma_225-Kaolack_225-1 225 GA SE 2.4 3.9

Mali_225-Sambangalou_225-1 225 GU SE 2.4 2.3

Bolgatanga_225-Ouaga Sud_225-1 225 GH BU 1.9 7.6

Elubo_225-Bingerville_225-1 225 GH CI 1.7 10.2

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 63/176

Fina

lver

sion

Table 9: Flows on interconnection – Asynchronous Peak 2022

Figure 27: Visualization of the active power flows - Peak 2022

From the load flow simulations at the peak it is observed that the East-West flowsare significant due to the importance of the exports of Nigeria. With most of thepower exported from Nigeria being sent to importing countries of Togo, Benin andBurkina, the North Core and the Ikeja West – Sakete interconnection are bothloaded at around 50% of their thermal capacity.

On the Western part of the region, the exporting country of Guinea is sendingmainly its power through CLSG to the countries of Sierra Leone and Liberia whichlack generation capacities to satisfy their peak load economically.

A general observation made is that most of the interconnections lines are nothighly loaded at the peak which gives space for greater exchanges betweencountries in the case of emergency situations or unlikely scenarios. It has to benoticed that the dynamic simulations presented in the next section 3.2.4.highlights further limitations due to stability limits and that the conclusions can befar different.

It was seen that the voltages in Niger cannot be held to the operational limits inthe eastern part where long 132 kV lines are present. The increase of load in theeast creates a voltage drop in Zinder which is below .95 p.u. In the short-termhorizon, these voltage problems were solved by the addition of capacitor banksin areas radially connected by long 132 kV lines.

Considering the short time period up to 2022 and the number of lines alreadydecided and needing to be built, the application of the N-1 criterion is not realisticand was not applied to the 2022 network. The list of problematic N-1 equipmentis shown here forth. Only the contingencies and overloads having a regionalinterest are shown in the table below. These identified weaknesses are furtheranalyzed and detailed in the dynamic analysis (next section).

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 64/176

Fina

lver

sion

Contingency Overload

CLSG Voltage collapse at the extremity which became radiallyconnected

Ikeja West 330 kV – Sakete 330 kV - 1 Voltage collapse in Lagos region

Volta 330 kV – Asogli 330 kV - 1 Overload of double circuit Davié – Lomé 161 kV

Ouaga South East 225 kV – Ouaga Sud 225 kV - 1 Overload of Patte D’Oie-Ouaga South East 132 kV

Table 10: List of problematic contingencies – Peak 2022

Figure 28: Non-secure N-1 contingencies - 2022 static peak scenario

3.2.3.2. SYNCHRONOUS OFF-PEAK 2022

The results of the synchronous off-peak scenario which was studied is shownhereunder. Table 11 shows the balances of each country in this scenario. Thesesbalances are representative of the economic dispatch and exchanges that resultfrom the optimization.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 65/176

Fina

lver

sion

Country Off-peak Balance

BURKINA - 1 MW

CIV 29 MW

GAMBIE 18 MW

GHANA 32 MW

GUINEE 264 MW

GUINEE BISSAU 24 MW

LIBERIA - 2 MW

MALI - 103 MW

NIGER 134 MW

NIGERIA -252 MW

SENEGAL 49 MW



Table 11: Country balance - Off-peak 2022

In the off-peak case studied here, renewable infeed is considered to be of 100%for the concerns of PV units and wind turbines. This scenario allows to identify ifthe totality of possible infeed that can be evacuated through the grid.

The flows on the interconnection lines are shown in the table below.


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)

Dawa_330-Davié_330-1 330 GH TB 131.5 20.3

Zabori_330-Birnin Kebbi_330-1 330 NR NI 88.7 28.4


Sakete_330-Ikeja West_330-1 330 TB NI 77.1 12.6

Siguiri_225-Sanakoroba_225-1 225 GU MA 70 22.4

Siguiri_225-Sanakoroba_225-2 225 GU MA 70 22.4

Bobo_330-Bolgatanga_330-1 330 BU GH 66.2 15.3


Bingerville_225-Elubo_225-1 225 CI GH 63.6 20.2

Sikasso_330-Bobo_330-1 330 MA BU 56.1 12.6




Yekepa_225-Man_225-1 225 LI CI 42.1 12.4

Yekepa_225-Man_225-2 225 LI CI 42.1 12.4

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 66/176

Fina

lver

sion


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)



Bakel_225-Kayes_225-1 225 SE MA 28.7 14.1

Aflao Ghana_161-Lomé (Aflao) 1_161-1 161 GH TB 26.4 23.2

Bawku_161-Cinkassé_161-1 161 GH TB 23.5 22.2

Linsan_225-Kamakwie_225-1 225 GU SL 21 7.7



Tambacounda_225-Kayes_225-1 225 SE MA 17.6 5.4


Ouaga Est_330-Goroubanda_330-1 330 BU NR 17 34.5

Ouaga Est_330-Goroubanda_330-2 330 BU NR 17 34.5

Salthinho_225-Boke_225-1 225 GB GU 16.5 7.2



Dosso_132-Birnin Kebbi_132-1 132 NR NI 13.3 12.7



Mano_225-Kenema_225-1 225 LI SL 6.7 5.6

Mano_225-Kenema_225-2 225 LI SL 6.7 5.6

Malanville_330-Zabori_330-1 330 TB NR 6.7 6.8

Tanaf_225-Soma_225-1 225 SE GA 2.7 3.3


Table 12: Flows on interconnection - Off-peak 2022

Table 13 shows the contingencies which are problematic in the 2022 off-peakcase. It can be noted that the contingency shown here in the table is linked to theproduction of Asogli 2 and that changing the dispatch of this one could removethe observed overload.

Contingency Overload

Asogli 330 kV -Dawa 330 kV - 1 Overload of Akosombo – Lomé Aflao 161 kV

Table 13: List of problematic contingencies – Off-peak 2022

3.2.4. Dynamic studies

3.2.4.1. SMALL SIGNAL STABILITY

This section presents the results of the small-signal stability analysis of the WAPPinterconnected network for target-year 2022, at peak and off-peak load conditions.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 67/176

Fina

lver

sion

3.2.4.1.1. Peak Load - Simulation resultsStarting from the initial load flow solution (base case), the eigenvalues of thenetwork in normal operation conditions are computed.

The foreseen PSS scheme, as communicated by the Client, provides sufficientdamping for the vast majority of modes, except for one interarea mode and fewlocal modes, presented in Table 14. Modes with a damping ratio higher than 5%are not included in the table.

Mode ID Eigenvalue DampingRatio [%]

Frequency[Hz] Participating plants

230 -0.204 + 10.8434j 1.8807 1.7258 Boutubre (CIV)

231 -0.204 + 10.8434j 1.8807 1.7258 Boutubre (CIV)

228 -0.2102 + 10.5608j 1.9901 1.6808 Boutubre, Gribo Popo (CIV)

229 -0.2153 + 10.65j 2.0212 1.695 Gribo Popo (CIV)

280 -0.0384 + 1.7201j 2.232 0.2738 Interarea: Nigeria/Niger vs. WAPP

Table 14: List of modes with damping ration below 5% - 2022 peak

A badly dampened interarea mode of 0.27 Hz is detected at peak load condition.The remaining local modes in Côte d’Ivoire can easily be solved by adding a PSS.

The current eastern synchronous block oscillates against the rest of the WAPP.The mode shape is shown in Figure 29

2022 Peak – initial case – Results of Small Signal Stability analysis

Interarea Mode NI/NR/TB vs WAPP Eigenvalues

Figure 29: Results of Small Signal Stability analysis - 2022 peak.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 68/176

Fina

lver

sion

The insufficient damping of the interarea mode causes power oscillations ofincreasing magnitude following a small variation of the operating point. The unitsat the extremities of the WAPP system, especially hydro ones, are subject to thelargest power oscillations. Figure 30 shows the speed of the units of Manantali(Mali) and Egbin 1 (Nigeria) following the loss of a unit in Egbin 2 (Nigeria).

Figure 30: Machine speed of Manantali (MA) and Egbin 2 (NI) following loss of one unit of Egbin 2 – 2022 peak.

In the following sections of this document, the abbreviation (Rx) will be used torefer to the set of recommendations that should be implemented by 2022.

The damping of the interarea mode is improved by reinforcing the network. Withboth reinforcements R2 and R4 in place, the interarea mode results damped to4.94% at peak load, close to the target but still insufficient.

However, improving the damping by additional reinforcements proved to beinefficient as too many investments would be required by 2022. Therefore, theConsultant recommends to expressly tune PSS of some large units at theextremities of the WAPP system (e.g. in Guinea / Mali on one side and in Nigeriaon the other side) to improve the damping of the critical interarea mode to a valueabove 6% (R1).

3.2.4.1.2. Off-peak Load - Simulation ResultsOff-peak load conditions are less challenging in terms of small signal stability. Theresults of the eigenvalues computation with the reinforcements in place arereported in Table 15 and Figure 31. The interarea mode is well dampened at off-peak load.

100 110 120 130 140 150 160 170 180 190

49.88

49.89

49.90

49.91

49.92

49.93

49.94

49.95

49.96

49.97

49.98

49.99

50.00

s

Hz

[F_EGBIN2G1_NI] MACHINE : MANAN11A SPEED Unit : Hz[F_EGBIN2G1_NI] MACHINE : 16000 1 SPEED Unit : Hz

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 69/176

Fina

lver

sion

Mode ID Eigenvalue DampingRatio [%]

Frequency[Hz] Participating plants

35 -0.2138 + 10.665j 2.0045 1.6974 Boutubre, Gribo Popo (CIV)

3 -0.8124 + 15.7435j 5.1532 2.5057 Kaduna (NI)

1 -0.7842 + 14.9396j 5.2416 2.3777 Azura (NI)

106 -0.1815 + 3.4553j 5.2464 0.5499 Interarea: Nigeria/Niger vs. WAPP

Table 15: List of modes with damping ration below 5.5% - 2022 off-peak

2022 Off-Peak – reinforced R2, R4 and R5– Results of Small Signal Stability analysis

Interarea Mode NI/NR/TB vs WAPP Eigenvalues

Figure 31: Results of Small Signal Stability analysis - 2022 off-peak.

3.2.4.2. DYNAMIC SECURITY ANALAYSIS

The objective of the dynamic security analyses is to verify the ability of the systemto endure faults in the transmission network without loss of synchronism ofgenerating units and other destabilizing phenomena such as voltage collapses.

The transmission system at 2022 presents two critical interfaces, correspondingapproximately to the borders between the current synchronous blocks. Theseinterfaces are illustrated in Figure 40 and are characterized by the followingcriticalities:

· Critical Interface 1 - Nigeria / Niger with the rest of WAPP: about half of theexported power from Nigeria (670 MW total in peak base case) is exportedalong the southern corridor through a single circuit 330 kV transmission line.On the other hand, the northern corridor connects two weak parts of BurkinaFaso and Niger.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 70/176

Fina

lver

sion

· Critical Interface 2 - Central WAPP with western WAPP: while the powertransferred through the interface is not significantly high, the two blocks will beinterconnected by only two circuits by 2022.

Three-phase short circuits cleared in 100 ms base time are simulated oninterconnection lines at these interfaces and on other relevant branches. The fullDSA methodology is presented in Annex.

3.2.4.2.1. Peak Load – Simulations ResultsCritical Interface 1 – Block C with the rest of WAPP

Faults cleared in 100 ms by tripping the faulted lines are simulated on the singlecircuit of the NI-TB interconnection and on one circuit of the North Coreinterconnection.

Losing the NI-TB interconnection causes a split of the system, a coherent groupof machines in Nigeria, Niger and Burkina will lose synchronism with the machinesof the rest of WAPP, as shown in Figure 32.

The split is caused by the redistribution of power flows following the loss of theNI-TB interconnection. The exported power from Nigeria is forced to pass entirelythrough the North Core causing angular and voltage instability (in Burkina Fasoand Niger).

2022 Peak – initial case – loss of NI-TB interconnection

Machine angle [deg] Voltage at Ouagadougou [p.u.]

Figure 32: Voltage and angle transients the following loss of NI-TB interconnection - 2022 peak initial case.

Following the same pattern, voltage instability is detected also following the lossof one circuit of the North Core. The flow on the remaining circuit and on the NI-TB interconnection increases, violating voltage stability limits and causing voltagecollapses in Burkina Faso and Niger. Eliminating these instabilities requires:

· Increasing the reactive power compensation in Burkina by adding a 100 MVArSVC at Ouagadougou substation (R5);

· Installing a Special Protection Scheme (SPS) to allow the expected energyexchanges between Nigeria and the rest of WAPP (R4). The flows beingexported from Nigeria would then be reduced to less than 350 MW in case ofcritical contingencies on the interface 1. In normal operation, higher exchangescould be maintained.

49.0 49.5 50.0 50.5 51.0 51.5 52.0 52.5 53.0 53.5 54.0 54.5 55.0 55.5

-50

-0

50

100

150

s

deg

[TS_08-A] MACHINE : 16000 1 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : MANAN11A ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : AKOSOM_1 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : MA_GLE1G ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : 4BAGRE16 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : KANDADG1 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : CIPR5-1 ANGULAR POSITION Unit : deg

49.0 49.5 50.0 50.5 51.0 51.5 52.0 52.5 53.0 53.5 54.0 54.5 55.0 55.50.050.100.150.200.250.300.350.400.450.500.550.600.650.700.750.800.850.900.951.001.05

s

[TS_08-A] VOLTAGE AT NODE OUAGA_E2 Unit : p.u.[TS_08-A] VOLTAGE AT NODE GOROUB02 Unit : p.u.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 71/176

Fina

lver

sion

In these conditions, the response of the system to the loss of the NI-TBinterconnection (worst case) is satisfactory, as presented in Figure 33.

2022 Peak – Reinforcements R4 and R5 – loss of NI-TB interconnection

Machine angle [deg] Voltage at Ouagadougou [p.u.]

Figure 33: Voltage and angle transients the following loss of NI-TB interconnection, 2022 peak with R4 and R5.

Critical Interface 2 - Central with western WAPP

The interface between Côte d’Ivoire and the western part of WAPP is composedon two single circuit interconnections. Losing any one of these interconnectionsleads to instabilities. Figure 34 shows the response of the system to the loss ofthe Sikasso (Mali) – Ferke (Côte d’Ivoire) interconnection.

2022 Peak – Reinforcements R4 – R5 – Loss of MA - CIV interconnection

Machine speed [Hz] Machine angle [deg]

Figure 34: Machine speed and angular transients following loss of MA - CIV interconnection, 2022 peak with R4 and R5.

Unstable oscillations are observed due to the excitation of the interarea mode.These undamped oscillations leads to voltage collapse at the border betweenCôte d’Ivoire and Liberia and along the CLSG route. Afterwards, the increasingangular deviation would end up in loss of synchronism and splitting of the system.

50 55 60 65 70 75 80-40

-30

-20

-10

-0

10

20

30

40

50

s

deg

[TS_08-A] MACHINE : 16000 1 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : MANAN11A ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : AKOSOM_1 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : MA_GLE1G ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : 4BAGRE16 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : KANDADG1 ANGULAR POSITION Unit : deg[TS_08-A] MACHINE : CIPR5-1 ANGULAR POSITION Unit : deg

50 55 60 65 70 75 80

0.98

0.99

1.00

1.01

1.02

1.03

s

p.u.

[TS_08-A] VOLTAGE AT NODE OUAGASE3 Unit : p.u.[TS_08-A] VOLTAGE AT NODE GOROUB02 Unit : p.u.

50 55 60 65 70 75 80 85 90 95 100

49.6

49.8

50.0

50.2

50.4

s

Hz

[TS_05-A] MACHINE : 16000 1 SPEED Unit : Hz[TS_05-A] MACHINE : MANAN11A SPEED Unit : Hz[TS_05-A] MACHINE : AKOSOM_1 SPEED Unit : Hz[TS_05-A] MACHINE : KOSSD_G1 SPEED Unit : Hz[TS_05-A] MACHINE : BUYOG1 SPEED Unit : Hz

50 55 60 65 70 75 80 85 90 95 100

-150

-100

-50

-0

50

s

deg

[TS_05-A] MACHINE : 16000 1 ANGULAR POSITION Unit : deg[TS_05-A] MACHINE : MANAN11A ANGULAR POSITION Unit : deg[TS_05-A] MACHINE : AKOSOM_1 ANGULAR POSITION Unit : deg[TS_05-A] MACHINE : BUYOG1 ANGULAR POSITION Unit : deg

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 72/176

Fina

lver

sion

Figure 35 shows the response of the system to the loss of the otherinterconnection, between Man (Côte d’Ivoire) and Yekepa (Liberia), part of theCLSG project.

2022 Peak – Reinforcements R4 and R5 – Loss of LI - CIV interconnection


Figure 35: Machine speed and angular transients following loss of MA - CIV interconnection, 2022 peak with R4 and R5.

In this case, it can be observed how the oscillations causes loss of synchronismin the system.

These unstable operating conditions can be mitigated by increasing the dampingof the interarea mode (R1) and by anticipating the investment of the 330 kVinterconnection between Sikasso (Mali), Bobo (Burkina Faso) and Bolgatanga(Ghana) (R2-A).

This investment also brings the added benefit of improving the dynamic stabilityof the North Core being its continuation. Figure 36 shows the satisfactoryresponse of the system with the reinforcements in place.

2022 Peak – Reinforcements R4, R5 and R2-A – Loss of LI - CIV interconnection


Figure 36: Machine speed and angular transients following loss of MA - CIV interconnection, 2022 peak with R4, R5 andR2-A.

50 55 60 65 70 75 80 85 90 95 100 105

48.8

49.0

49.2

49.4

49.6

49.8

50.0

50.2

50.4

50.6

s

Hz

[TS_02-A] MACHINE : 16000 1 SPEED Unit : Hz[TS_02-A] MACHINE : MANAN11A SPEED Unit : Hz[TS_02-A] MACHINE : AKOSOM_1 SPEED Unit : Hz[TS_02-A] MACHINE : BUYOG1 SPEED Unit : Hz

50 55 60 65 70 75 80 85 90 95 100 105

-1500

-1000

-500

-0

s

deg


50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150

49.94

49.96

49.98

50.00

50.02

50.04

50.06

50.08

50.10

s

Hz

[TS_02-A] MACHINE : 16000 1 SPEED Unit : Hz[TS_02-A] MACHINE : MANAN11A SPEED Unit : Hz[TS_02-A] MACHINE : AKOSOM_1 SPEED Unit : Hz[TS_02-A] MACHINE : BUYOG1 SPEED Unit : Hz

50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 150

-50

-40

-30

-20

-10

-0

10

20

30

40

s

deg


WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 73/176

Fina

lver

sion

Other results of the DSA analysis

The reinforced system has been tested against the loss of several other keytransmission lines. The list of selected incidents and the results of the analysisare reported in Annex.

The analyses show voltage stability issues along the CLSG transmission line.When the single-circuit interconnection between Linsan (GU) and Kamakwie (SL)or between Kamakwie (SL) and Yiben (SL) is tripped, voltage collapses aredetected in several substations in Liberia and Sierra Leone, as presented inFigure 37.

Figure 37: Voltage transients along CLSG following the tripping of the Linsan (GU) – Kamakwie (SL) line – 2022 peak R4,R5 and R2-A.

To address this issue, the Consultant recommends to:

· Anticipate the second circuit of CLSG project (building CLSG directly with 2circuits), interconnecting Guinea to Cote d’Ivoire, in order to ensure N-1security on that border (R2-B);

These recommendations have to be intended as minimal remedial actions.

A dynamic security analysis has been carried out on the reinforced networkimplementing a dynamic load model with 40% of rotating loads. For the loss ofone circuit of the BU-NR North Core interconnection, voltage collapses areobserved in Burkina Faso and Niger. Additional dynamic voltage support isrequired. The best option is to install a 200 MVAr SVC at Salkadama (Niger) (R5).This substation is suitable because it’s connected to the rest of the system throughvery long 330 kV AC lines and it might serve as connection point for futureinterconnections, maximizing the technical benefits of the SVC. The results withand without the SVC are presented in Figure 38.

49 50 51 52 53 54 55 56 57 58 59 60 610.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

s

p.u.

[TS_01-A] VOLTAGE AT NODE YIBEN_03 Unit : p.u.[TS_01-A] VOLTAGE AT NODE BUMBUN03 Unit : p.u.[TS_01-A] VOLTAGE AT NODE MANO__03 Unit : p.u.[TS_01-A] VOLTAGE AT NODE MONROV03 Unit : p.u.[TS_01-A] VOLTAGE AT NODE YEKEPA03 Unit : p.u.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 74/176

Fina

lver

sion

2022 Peak – Reinforcements R4, R5 and R2 – Loss of BU - NR interconnection

Voltage transients in BU and NR without new SVC [p.u] Voltage transients in BU and NR with new SVC [p.u]

Figure 38: Voltage transients in BU and NR following loss of one circuit of the BU - NR interconnection, 2022 peak withR4, R5 and R2.

Other relevant results are the following:

· For a 100 ms fault on the 225 kV single-circuit line connecting Man (Coted’Ivoire) and Yekepa (Liberia), the interarea mode is excited causing voltageoscillation at Ferke (Côte d’Ivoire);

· For a 100 ms fault on the 132 kV single-circuit line connecting Gazaou (Niger)and Katsina (Nigeria), the substations from Maradi (NR) downwards will findthemselves at the end of a long feeder. Localized voltage under-voltages willtake place. The phenomenon has no impact on the regional operation of theWAPP system.

3.2.4.2.2. Off-Peak Load – Simulations ResultsAt off-peak load, the results do not show signs of transient instability except whathas already been detected for peak load. The results are reported in detail inAnnex.

49.0 49.5 50.0 50.5 51.0 51.5 52.0 52.5 53.0 53.5 54.0 54.5 55.00.000.050.100.150.200.250.300.350.400.450.500.550.600.650.700.750.800.850.900.951.001.05

s

p.u.

[TS_03-A] VOLTAGE AT NODE SALKAD02 Unit : p.u.[TS_03-A] VOLTAGE AT NODE GOROUB02 Unit : p.u.[TS_03-A] VOLTAGE AT NODE OUAGAE02 Unit : p.u.[TS_03-A] VOLTAGE AT NODE OUAGASE3 Unit : p.u.

49.0 49.5 50.0 50.5 51.0 51.5 52.0 52.5 53.0 53.5 54.0 54.5 55.0

0.150.200.250.300.350.400.450.500.550.600.650.700.750.800.850.900.951.001.05

s

p.u.

[TS_03-A] VOLTAGE AT NODE SALKAD02 Unit : p.u.[TS_03-A] VOLTAGE AT NODE GOROUB02 Unit : p.u.[TS_03-A] VOLTAGE AT NODE OUAGAE02 Unit : p.u.[TS_03-A] VOLTAGE AT NODE OUAGASE3 Unit : p.u.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 75/176

Fina

lver

sion

3.2.4.3. FREQUENCY STABILITY

The objective of the frequency stability analyses is to verify the ability of thesystem to endure transient phenomena caused by active power unbalances suchas the loss of the large generators and loads in various zones of the WAPPsystem. The methodology is detailed in Annex.

3.2.4.3.1. Peak load – Simulation ResultsThe results of the frequency stability simulations for the loss of the followinggeneration units and large loads is detailed in Table 16. Fault time is at 50seconds.

CountryType

(unit /load)

Plant / LoadLost

Power(MW)

Machines speed

Stable CommentsMin (Hz) Time of

min (s) Max (Hz) Time ofmax (s)

NI unit Egbin 2 285 49.886 52.505 50.000 50.338 yes damped oscillations

CIV unit Soubre 3 87 49.926 50.233 50.004 50.646 yes excited oscillations

GH unit Akosombo 1 140 49.885 419.796 50.057 421.534 yes excited oscillations

GU unit Souapiti 112.5 49.864 50.526 50.062 52.011 yes excited oscillations

MA unit Albatros 92 49.897 50.262 50.030 52.287 yes excited oscillations

TB unit Maria Gleta 100 49.937 143.292 50.012 172.867 yes excited oscillations

BU unit Ouagadougou 50 49.925 50.153 50.033 50.456 yes damped oscillations

NI load Benin 340 50.000 50.187 50.081 52.489 yes damped oscillations

SL load Bumbuna 174 49.912 50.411 50.283 50.150 yes damped oscillations

Table 16: Results of frequency stability analysis - 2022 peak with R2, R3 and R4

All simulated incidents resulted in acceptable frequency transients. The inertia ofthe interconnected WAPP system and the allocated operating reserve is sufficientto prevent excessive frequency drops and overshoots. For instance, the rate ofchange of frequency amount to approximately 0.04 Hz/s.

However, the loss of certain units excites the interarea mode causing frequencyoscillations. The worst cases are presented in red in the table above. Figure 39shows the speed of one machine in Nigeria and one in Mali oscillating in phasewith each other. The frequency of the oscillations is 0.28 Hz, in line with theobserved interarea mode.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 76/176

Fina

lver

sion

2022 Peak – Reinforcements R2, R3 and R4 and R2-A – Loss of unit Akosombo 1 at T = 50s

System frequency [Hz] Machine speed [Hz]

Machine speed [Hz] – Zoom on first 20 s after fault Machine speed [Hz] – Zoom on oscillations

Figure 39: System response to the loss of one unit at Akosombo at T = 50s, 2022 peak with R4, R5 and R2.

It is also observed that the worst cases involve the southern corridor.Nevertheless, reinforcing the 330 kV backbone has proved to be inefficient. Thus,the preferred solution is to accurately dampen the interarea modes of the system.

3.2.4.3.2. Off-Peak load – Simulations resultsAt off-peak load the frequency transients remain within acceptable ranges, asshown in Table 17 (fault time at 50s). The inertia and operating reserve of theinterconnected WAPP network are sufficient at off-peak conditions;

50 100 150 200 250 300 350 400

49.955

49.960

49.965

49.970

49.975

49.980

49.985

49.990

49.995

50.000

s

Hz

[FS_03] NODE FREQUENCY MANANT1G Unit : Hz

50 100 150 200 250 300 350 400

49.89

49.90

49.91

49.92

49.93

49.94

49.95

49.96

49.97

49.98

49.99

50.00

50.01

50.02

50.03

50.04

50.05

s

Hz

[FS_03] MACHINE : 16000 1 SPEED Unit : Hz[FS_03] MACHINE : MANAN11A SPEED Unit : Hz

49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 7049.90

49.91

49.92

49.93

49.94

49.95

49.96

49.97

49.98

49.99

50.00

50.01

50.02

s

Hz


215 220 225 230 235 240 245 250

49.90

49.92

49.94

49.96

49.98

50.00

50.02

50.04

s

Hz


WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 77/176

Fina

lver

sion

Country

Type

Plant / Load

Lost Machines speed

Stable Comments(unit /load)

Power(MW)

Min(Hz)

Time ofmin (s)

Max(Hz)

Time ofmax (s)

NI unit Egbin 2 285 49.797 51.491 50.001 50.181 yes

CIV unit Soubre 1 61 49.938 52.154 50.008 51.165 yes

GH unit Akosombo 1 140 49.861 52.331 50.000 50.016 yes

GU unit Souapiti 108 49.715 50.321 50.023 50.717 yesrequired voltagesupport at Manantali,operate always atleast 2 hydro units

MA unit Manantali 5 38 49.951 50.158 50.006 50.973 yes

NI load Benin 177 49.999 50.130 50.083 53.081 yes

Table 17: Results of frequency stability analysis - 2022 off-peak with R2, R4 and R5

The following operational recommendations are drawn from the off-peak results:

· Dynamic voltage support is a critical issue when a generation unit is lost at off-peak. To this end, it is recommended that the hydro power plant of Manantalishould be operated with at least two units in operation.

· The Mauritanian system is subject to voltage collapses following the loss of agenerating units in several locations in the WAPP.

3.2.5. Technical operation of the network in 2022It should be noted that for the study year 2022, the network is not yet N-1compliant.

In 2022, the network studies have highlighted steady-state security issuesconcerning the import and export of high amounts of power from and to Nigeria.As seen from the economic study, Nigeria should export up to 700 MW at theevening peak. Considering that in 2022, only two evacuation corridors are present(the double-circuit North Core going through North of Nigeria, through Niger andto Burkina Faso; and the single circuit 330 kV line from Nigeria to Benin), losingone of these two corridors causes steady-state problems to continue evacuatingthe 700 MW from Nigeria to the rest of the WAPP. The issue is confirmed by thedynamic analyses.

Considering the economic exchanges and the investments from the economicanalysis, it is seen that only one circuit is necessary for the CLSG line in 2022.However, the investment in the double circuit is justified by 2022 for the systemstability and should thus be a priority.

Several challenges to the dynamic stability of the synchronous WAPP systemhave been identified. The long-distance power flows will subject the system tointerarea modes (large groups of generation units swinging in opposition to eachother following small disturbances on the network). The existing Power SystemStabilizer scheme dampens most of these dangerous modes, except for oneinterarea mode at 0.27 Hz between the current eastern part of WAPP (fromNigeria to Ghana) against the rest of WAPP.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 78/176

Fina

lver

sion

The recently finalized “WAPP synchronization study” already recommendedinstalling additional PSS but this measure should be further extended before thesynchronisation of the entire WAPP system. Dedicated tuning of PSS is requiredto damp interarea oscillations. In particular, the Consultant recommends toexpressly tune PSS on large units at the extremities of the WAPP system toimprove the damping of a critical 0.27 Hz interarea mode (R1) and reducedangerous oscillations.

Stability issues have been detected at the interfaces between the existingsynchronous blocks.

Figure 40: Scheme of the interconnected WAPP system by 2022.

These interfaces are characterized by the following criticalities:

· Critical Interface 1 - Nigeria / Niger with the rest of WAPP: The economicstudy demonstrated the need of a strong interface to allow energy exchangesin both directions (up to 700 MW), importing solar energy during the day andexporting gas-based energy at peak load conditions. With such level of energyexchanges, voltage stability limits of the interconnections are violated whenone of them is tripped. In the short term, reinforcing this interface might not befeasible. It is recommended to install a Special Protection Scheme (SPS) inorder to reduce the total exported power from Nigeria to 350 MW in the caseof a single line contingency on this interface, until cross-border transfercapacity is reinforced (R4). Additional dynamic reactive power compensationshould be implemented in Burkina Faso and Niger as well (R5).

· Critical Interface 2 - Central WAPP with western WAPP: the central part ofWAPP is expected to be interconnected with the western part through twosingle circuit 225 kV transmission lines: the initial part of the CLSG project andthe single circuit interconnection between Cote d’Ivoire and Mali. Also in thiscase, the border is not secure even with limited cross-border flows, as losingone key transmission lines nearby would cause instabilities. Two investmentscan be anticipated:- R2-A: The 330 kV interconnection between Sikasso (Mali), Bobo (Burkina

Faso) and Bolgatanga (Ghana);- R2-B: The second circuits of the CLSG project, this CLSG line should be

directly built with the double circuit to ensure the stability of the system.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 79/176

Fina

lver

sion

The response of the interconnected system to frequency transients is sufficientin all tested cases. The inertia of the system is high enough to avoid triggeringany UFLS threshold.

In terms of voltage stability, critical areas are detected: the eastern side ofBurkina Faso and the CLSG sections in Liberia and Sierra Leone.

In addition to the recommendations of the synchronization study, it isrecommended to evaluate further measures of voltage support (R5) such asinstalling two additional SVCs: 100 MVAr at Ouagadougou (Burkina Faso) and200 MVAr at Salkadama (Niger).

Based on the dynamic simulations, the following set of remedial actions isrecommended to achieve sufficient stability of the interconnected WAPP systemat the short term:

ID Recommendations

R1Tune PSS of some large units at the extremities of the WAPP system to improve thedamping of a critical 0.27 Hz interarea mode between eastern WAPP and the rest ofWAPP

R2

Reinforcing the Central / Western WAPP interconnections by anticipating twoinvestments:

a. 330 kV interconnection between Sikasso (Mali), Bobo (Burkina Faso) andBolgatanga (Ghana);

b. Second circuit of the CLSG project;

R3 Update the Operational Manual of WAPP to ensure a smooth synchronization andharmonize the defence action plan (UFLS, interconnection protections)

R4 Installing a Special Protection Scheme (SPS) to allow the expected energy exchangesbetween Nigeria and the rest of WAPP

R5 Improve dynamic voltage compensation by adding one SVC at Ouagadougou (BU) andone at Salkadama (NR)

Table 18: List of recommendations to improve dynamic stability at the short term.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 80/176

Fina

lver

sion

3.3. Mid term development plan - 2025

3.3.1. Increasing the security of the systemIn the medium term, at the horizon of 2025, the WAPP’s electricity network isbecoming more and more interconnected with the increase of the demand leveland of exchanges between countries.

In order to satisfy the security of supply, the WAPP network is now moving to aN-1 secure network for which the loss of any high voltage equipment in thenetwork should not cause any major problem on the grid.

From the 2022 dynamic analysis, several weak points of the system werehighlighted. These different weak links should be reinforced by 2025 in order tooperate the system within the acceptable security limits under any singlecontingency. The studies realized for the target year 2025 will have as mainconclusions of confirming the list of needed investments in order to properlyoperate the synchronous network under the economic exchanges and the definedsecurity criteria.

To achieve this challenging objective, the following high voltage projects shouldbe prioritized.

NEW DOUBLE CIRCUIT 225 KV GUINEA – CÔTE D’IVOIREINTERCONNECTION

With the arrival of the Morisanako hydro power plant of 100 MW which shouldeconomically be commissioned in 2025, the interconnection line between Côted’Ivoire and Guinea going through the Morisanako site should be put in service.This 380 km double circuit line will connect the 225 kV Boundiali substation (Côted’Ivoire) with the 225 kV Fomi substation in Guinea. The line will allow theevacuation of the power from the Morisanako site which comprises of the 100 MWhydro power plant and an additional 100 MW PV park which should becommissioned at the same time. Additionally, considering the potential of theNorthern part of Côte d’Ivoire, this line will allow the sharing of renewable energyfrom Côte d’Ivoire with hydro power from Guinea. Finally, as mentioned in thedynamic analysis of year 2022, this link brings big benefits in the system stabilitytowards an N-1 secure network.

NEW MEDIAN BACKBONE

This double circuit 330 kV line is to be built between Nigeria, Benin, Togo, Ghanaand Côte d’Ivoire. The line will connect at the existing 330 kV substation of Shiroro(Nigeria), to the existing 330 kV substation of Kainji (Nigeria), to the new 330 kVsubstation in Parakou (Benin), the new 330 kV substation in Kara (Togo), the new330kV substation at Yendi (Ghana), the existing 330 kV substation in Tamale(Ghana) and the new 330 kV substation in Ferkéssédougou (Côte d’Ivoire). Thetotal distance of this line is approximately 1350 km.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 81/176

Fina

lver

sion

This line is of crucial importance for the integration of renewable energy in thelong term. By 2033, the needs of Nigeria to import renewable energy from thenorth of Ghana, the north of Côte d’Ivoire, Mali and Burkina Faso shows theimportance of such a new corridor which links these regions together. Theconnection of the median backbone in Benin and Togo allows for the sharing ofrenewable potential in these regions whose load is less significant, andpossibilities of export are existent. The connection at Kainji proposes thepossibility of exporting the hydro power from Nigeria which is located in the Northand in the East (Mambilla). Finally, this interconnection offers the advantage ofreducing the flows on the North Core when Nigeria is importing a lot of power fromthe Northern countries, which facilitates the operation and security of the network.

The approximate path of this line and connection points is shown in the illustrationbelow

Figure 41: Median backbone connection points

The dynamic simulations for the 2022 horizon have shown that the maximumtransfer capacity of Nigeria to the rest of the WAPP is of around 350 MW undercontingency N-1 and without SPS. Considering the economic exchanges that areexpected in 2025 which account for maximum exports of more than 400 MW andmaximum imports of more than 800 MW, an additional interconnection fromNigeria to the rest of the WAPP is necessary. This interconnection will greatlyincrease the system stability and allow for a better sharing of resources betweenNigeria and the rest of the WAPP.

The increase of transfer capacity brought by the commissioning of thisinterconnection is evaluated in Table 19.

= −

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 82/176

Fina

lver

sion

Where TTC is the total transfer capacity we are evaluating, NTC is the totalpossible transfer under N-1 secure rule between country A and country B and theTRM is a reliability margin which is typically taken of around 10%.

Country A Country B TTC increase

WAPP Nigeria + 850 MW10

Nigeria WAPP + 700 MW

Table 19: Increase of TTC between Nigeria and WAPP

Dynamic analyses have confirmed the positive impact of the median backbone onthe stability of the system. The median backbone improves the transfer capacityon the border between Nigeria and the rest of WAPP, unlocking more import andexport potential (see 3.3.4.2)

NEW 225 kV DOUBLE CIRCUIT LINE LINSAN – KOUKOUTAMBA –MANANTALI

A new 225 kV interconnection is planned between Guinea and Mali in the 2025horizon. With the commissioning of the Koukoutamba hydro power plant of closeto 300 MW during the 2022-2025 period, a new interconnection line to export thispower is necessary. This new double circuit interconnection will connect at theexisting 225 kV substation in Linsan (Guinea) on one side and at the existing 225kV Manantali substation (Mali) on the other. A new substation will be created atKoukoutamba in order to evacuate the power from Koukoutamba on this line.Additionally, the path of this line should go through the future site of Boureyawhich will be commissioned in the long term. This line will then serve at theevacuation of both of these hydro power plants and increase the exportingcapabilities of Guinea to Mali.

NEW 225 kV DOUBLE CIRCUIT LINE LABÉ - KOUKOUTAMBA

From the generation master plan detailed in Chapter 2, it is seen that investmentsin Fetore (124 MW), Bonkon Diara (174 MW - 2025) and Grand Kinkon (291 MW- 2023) should be realized by 2033 and are economically justified. The connectionof these power plants was assumed to be done at the Labé substation. In orderto securely evacuate the power from these hydro power plants (total of 589 MW),the existing OMVG loop is not sufficient in N-1 situation. The commissioning of anew connection in Labé is thus necessary in order to respect the N-1 condition.

A new 225 kV double circuit of approximately 115 km is thus proposed from Labéto Koukoutamba. This line creates a direct link between the OMVG loop and theLinsan-Manantali interconnection between Guinea and Mali where the generationfrom Koukoutamba and Boureya are also injected. This line is proposed to becommissioned in 2025 at the same time as the Bonkon Diara project which willincrease the installed capacity at Labé from 291 MW (Grand Kinkon only) to 465MW (Grand Kinkon and Bonkon Diara).

10 These values were calculated from the static analysis at the 2033 horizon

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 83/176

Fina

lver

sion

Figure 42: Labé - Koukoutamba line

3.3.2. Modelling of the 2025 WAPP NetworkThe hypothesis and information that was used to create the 2025 model of theWAPP are described in the following sections.

3.3.2.1. NATIONAL REINFORCEMENTS

The list of national reinforcements that were done on the high voltage grid for eachcountry in 2025 is described in Annex.


The load levels which were modelled in the 2025 scenario is detailed in thisparagraph. The modelled load corresponds to the yearly peak load of everycountry. This scenario where every country observes their peak load at the sametime is a conservative way of analyzing the reinforcement needs on the grid. Theresulting active load levels are presented in Table 20. The same power factor asthe existing model was kept for this study year.

Country Peak Load 2025

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 84/176

Fina

lver

sion

BENIN 431 MW

BURKINA 613 MW

CIV 2434 MW

GAMBIE 173 MW

GHANA 3597 MW

GUINEE 666 MW


LIBERIA 218 MW

MALI 778 MW

NIGER 554 MW

NIGERIA 15000 MW11

SENEGAL 1356 MW

SIERRA LEONE 487 MW

TOGO 397 MW

TOTAL 26834 MW

Table 20: Load level – Peak 2025

3.3.2.3. DYNAMIC MODEL

The dynamic model of 2025 is derived from the 2022 one. The additionalgenerating units have been modelled using parameters of similar units (size andtype). Standard controllers have been implemented.

Each new hydro units have been equipped with a wide-range standard PSS. It isnoted that for transient stability analysis, the interarea mode has been artificiallydamped. A proper tuning should be carried out at a later stage.

A distribution type load model is adopted for the 2025 dynamic analyses.

3.3.3. Static StudiesSimilarly as for 2022, the asynchronous peak, in which every country experiencesits peak at the same time, has been modelled. The country balances resultingfrom the economic exchanges is presented in Table 21. Negative values indicateimport.

Country Peak Balance2025

BURKINA - 506 MW

11 The same load level as in 2022 was modelled with a modification of the import/export balance based on thesimulations of the economic optimization. The needed internal reinforcements by 2025 are based on theTCN master plan.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 85/176

Fina

lver

sion

CIV 34 MW

GAMBIE - 111 MW

GHANA 69 MW

GUINEE 1150 MW

GUINEE BISSAU -87 MW

LIBERIA - 131 MW

MALI - 380 MW

NIGER - 149 MW

NIGERIA 451 MW

SENEGAL -25 MW



Table 21: Country balance – Peak 2025

It is seen that in the evening peak of this target year 2025, Burkina Faso isimporting up to almost 80% of its load. Also Mali is importing a lot. On the otherhand, Guinea is exporting a large part of their hydro generation which is clearly inexcess compared to their national load. Nigeria is also exporting more than 450MW thanks to the gas availabilities.

The flows on the interconnection lines are shown here in Table 29.


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)


Boureya_225-Manantali_225-1 225 GU MA 152.1 45.5


Kainji_330-Parakou_330-1 330 NI TB 129.6 17.1


Dawa_330-Davié_330-1 330 GH TB 128.6 12.6




Morisanako_225-Boundia_225-1 225 GU CI 89.5 26.9






Kaolack_225-Soma_225-1 225 SE GA 76.3 23.4

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 86/176

Fina

lver

sion


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)

Birnin Kebbi_330-Zabori_330-1 330 NI NR 75.3 11.5






Kara_330-Yendi_330-1 330 TB GH 61.6 8.1

Kara_330-Yendi_330-2 330 TB GH 61.6 8.1




Kayes_225-Bakel_225-1 225 MA SE 50.3 23.3





Tanaf_225-Soma_225-1 225 SE GA 36.6 11.2

Kenema_225-Mano_225-1 225 SL LI 32 9.7

Kenema_225-Mano_225-2 225 SL LI 32 9.7




Yekepa_225-Man_225-1 225 LI CI 11.2 6.3

Yekepa_225-Man_225-2 225 LI CI 11.2 6.3

Asiekpe PST_161-Lomé (Aflao) 1_161-1 161 GH TB 10 10.9

Cinkassé_161-Bawku_161-1 161 TB GH 9.4 5.3





Table 22: Flows on interconnection - Peak 2025

In 2025, the security analysis has shown that no contingency is problematic onthe regional high voltage grid. The list below shows the contingencies which cancause a problem on the national grid. These contingencies do not have a regionalimpact and should be treated by the national master plans of the countryimpacted. The observed N-1 problems should be treated on the national level.

Contingency Country Overload

Bondoukou 225 kV – Serebou 225 – 1 (NATIONAL) CI Voltage Collapse in the area of Bouna

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 87/176

Fina

lver

sion

Toulepleu 225 kV – Zagne 225 kV – 1 (NATIONAL) CI Voltage Collapse in the area of Mine Ity

Dueckoue 225 kV – Zagne 225 kV – 1 (NATIONAL) CI Voltage Collapse in the area of Mine Ity

Toulepleu 225 kV - Toulepleu 90 kV -1 (NATIONAL) CI Voltage Collapse in the area of Mine Ity

Table 23: List of problematic contingencies (NATIONAL) – Peak 2025

3.3.4. Dynamic studies

3.3.4.1. INTERAREA OSCILLATIONS

Regardless of the added network equipment, the interarea mode detected in 2022is still present in 2025. This is due to the commissioning of several new hydrounits in Guinea that contribute to the interarea oscillations.

It is recommended to implement PSS to the planned hydro units, properly tunedto dampen the interarea mode.

3.3.4.2. CRITICAL INTERFACE 1 - IMPACT OF THE MEDIAN BACKBONE

The addition of the median backbone greatly improve the transient stability of thesystem following a fault on the border between Nigeria and the rest of WAPP,unlocking higher transfer capacity levels. This impact is estimated in the followingsections.

3.3.4.2.1. Export from NigeriaAt peak load conditions with an export from Nigeria of 700 MW, without theMedian Backbone, losing the single circuit 330 kV interconnection between Ikeja(NI) and Sakete (TB) will results in the violation of the voltage stability limits of theNorth Core interconnection. With the Median Backbone in place, these limits arenot trespassed. Figure 43 shows the two situations.

2025 Peak – Export of 700 MW from Nigeria - loss of NI-TB interconnection - Voltage in BU and NR [p.u.]

WITHOUT Median Backbone WITH Median Backbone

Figure 43: Voltage transients following loss of NI-TB interconnection with and without median backbone- 2025 peak.

49 50 51 52 53 54 55 56 57

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

s

p.u.

[TS_08-A] VOLTAGE AT NODE OUAGA_E1 Unit : p.u.[TS_08-A] VOLTAGE AT NODE KANDADG2 Unit : p.u.

49 50 51 52 53 54 55 56 57 58 59 600.5

0.6

0.7

0.8

0.9

1.0

s

p.u.

[TS_08-A] VOLTAGE AT NODE OUAGA_E1 Unit : p.u.[TS_08-A] VOLTAGE AT NODE KANDADG2 Unit : p.u.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 88/176

Fina

lver

sion

Moreover, the median backbone enables higher export capacity, up to 900 MW.As shown in Figure 44, an export of 900 MW excites the interarea mode but theoscillations will dampen relatively quickly. For higher levels of export, theoscillations will be progressively amplified, leading to loss of stability.

2025 Peak – Export limits of Nigeria - loss of NI-TB interconnection – Machine speed [Hz.]

Export of 900 MW

Export of 950 MW

Export of 1000 MW

Figure 44: Machine speed transients following loss of NI-TB interconnection with median backbone for different exportlevels of Nigeria- 2025 peak.

60 80 100 120 140 160 180 200 220 240 260 280 300

49.8

49.9

50.0

50.1

50.2

s

Hz

[TS_08-A] MACHINE : MANAN11A SPEED Unit : Hz[TS_08-A] MACHINE : EGBIN2G1 SPEED Unit : Hz

60 80 100 120 140 160 180 200 220 240 260 280 300

49.8

49.9

50.0

50.1

50.2

s

Hz


60 80 100 120 140 160 180 200 220 240 260 280 300

49.8

49.9

50.0

50.1

50.2

s

Hz


WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 89/176

Fina

lver

sion

3.3.4.2.2. Import to NigeriaThe import capacity to Nigeria has been tested with and without medianbackbone. The results show a definite increase in dynamic performance with thebackbone in place. For instance, with an import level of 850 MW, losing one largeunit in Nigeria or the single circuit 330 kV interconnection between Nigeria andTogo-Benin would result in a stable response of the system, as shown in Figure45.

2025 Peak – Import of 850 MW to Nigeria -– Machine speed [Hz.]

Loss of NI-TB interconnection

Loss of one unit at Egbin (~190 MW of lost generation)

Figure 45: Machine speed transients with median backbone for 850 MW of import, under different contingency - 2025peak.

50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

49.85

49.90

49.95

50.00

50.05

50.10

50.15

s

Hz

[TS_08-A] MACHINE : EGBIN2G1 SPEED Unit : Hz[TS_08-A] MACHINE : MANAN11A SPEED Unit : Hz

50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 20049.985

49.990

49.995

50.000

50.005

50.010

50.015

50.020

s

Hz

[FS_01_v2] MACHINE : 16001 1 SPEED Unit : Hz[FS_01_v2] MACHINE : MANAN11A SPEED Unit : Hz

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 90/176

Fina

lver

sion

3.3.4.3. CRITICAL INTERFACE 2 – SECURITY VERIFICATION

The interface between Ivory Coast and the western part of WAPP poses securitychallenges in 2022. Different remedial actions have been proposed to eliminatethe risk.

The results of dynamic analyses have proven that this interface remains securein 2025. In fact, the border between Côte d’Ivoire and Guinea will be furtherstrengthened with the commissioning of a 225 kV interconnection line betweenBoundiali (CIV) and Fomi (GU) by 2025.

The simulation results for the loss of the Sikasso (Mali) – Ferke (Ivory Coast)interconnection (most critical fault in 2022) are presented in Figure 46.

2025 Peak – Loss of MA - CIV interconnection


Figure 46: Machine speed and angular transients following loss of MA - CIV interconnection, 2025 peak.

50 55 60 65 70 75 80 85 90 95 100

49.95

50.00

50.05

50.10

50.15

s

Hz

[TS_32-A] MACHINE : 16000 1 SPEED Unit : Hz[TS_32-A] MACHINE : MANAN11A SPEED Unit : Hz[TS_32-A] MACHINE : AKOSOM_1 SPEED Unit : Hz[TS_32-A] MACHINE : KOSSD_G1 SPEED Unit : Hz[TS_32-A] MACHINE : BUYOG1 SPEED Unit : Hz

50 55 60 65 70 75 80 85 90 95 100

30

40

50

60

70

80

s

deg


WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 91/176

Fina

lver

sion

3.3.5. Conclusions and recommendations for 2025As it is seen in the studies realized for the target year of 2025, the objective ofsatisfying the N-1 security rule over the whole high voltage network is verychallenging in 2025. Many high voltage investments are necessary in order tooperate under the defined security limits and cope with the load increase as wellas the increased exchanges.

The different recommendations presented in the results of the target year 2022are fully confirmed my 2025 simulations and should remain a priority in order tostabilize the grid.

The connection of Nigeria-Niger with the rest of WAPP and the maximum powertransfer of this section are significantly improved thanks to the Median Backbonefrom Cote d’Ivoire to Nigeria. The stability limits in contingency N-1 is increasedup to 900MW.

3.4. Long term development plan - 2033

3.4.1. Increasing the sharing of resourcesIn the long term, the integration of renewable energy has increased significantlyto a level of about 18 % in the energy mix. The intermittent and variable nature ofthese renewable sources makes for large variations in the exchanges that areobserved over a typical day. One good example of this phenomenon is the caseof Nigeria which export about 1.5 GW of power during the evening peak andimports more than 2 GW of power during times of high renewable infeed in theWAPP.

In order to allow for this increase of exchanges, the network will need to be furtherreinforced. Based on the results from the economic analysis giving the optimalexchanges between countries, the technical analysis will allow to determine thereinforcement needs in order to securely satisfy these exchanges. The objectiveof the analysis of this target year 2033 is to define the best structure for theWAPP interconnected network and to verify the ability of this network to operatewith an increasing share of renewable energy.

The following projects should be set as the priority for developing a stablestructure for the WAPP to operate their interconnected network considering theload increase, the increase of renewable penetration on the system and theincrease of the economic exchanges between countries.

WESTERN BACKBONE

Starting in 2025 with the great increase of available gas resources in Senegal andthe increase of the installed capacity from CCGTs, it is expected that Senegal willbecome an exporting country during times of the year. In the same way, the hydropotential of Guinea being so large, the country will export large amounts of hydropower over the year. Simultaneously, Mali, Burkina and the north of Côte d’Ivoireand Ghana will be exporters of large amounts of renewable energy at times ofhigh solar radiation.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 92/176

Fina

lver

sion

Considering this spreading of the resources, the sharing of power from these threeresources (gas from Senegal, Hydro from Guinea and RES from the Central-North) becomes primordial and a high voltage corridor can significantly increasethe security of the grid.

In order to evaluate the benefits of creating a high voltage backbone connectingthese different western countries, total transfer capacities were calculatedbetween these different countries. This total transfer capacity was calculatedunder the N-1 security rule and the following expression:

= −

Where TTC is the total transfer capacity we are evaluating, NTC is the totalpossible transfer under N-1 secure rule between country A and country B and theTRM is a reliability margin which is typically taken of around 10%.

Country A Country B TTC increase

Senegal Guinea + 650 MW12

Guinea Mali + 325 MW

Senegal Mali + 500 MW

Table 24: Increase of TTC with the 330 kV Western backbone

Considering these transfer capacities of the existing grid, it was concluded thatthe following substation could be good candidates for the sharing of resourcesthrough this high voltage 330 kV line. The substation of Tobène where theinterconnection with Mauritania is planned as well as a possible futureinterconnection with Morocco. Furthermore, this substation is at the crossroad ofmany of the existing 225 kV line (to Sakal, Kounoune, Touba, Taiba …) and thusseems as a good injection point to a higher voltage level. Additionally, its proximityto the sea makes it a good candidate for the future connection points of newCCGTs.

The substation of Linsan also makes for a good candidate substation in Guineadue to the large capacity of hydro power plant located close to this substation.Additionally, it can be noted that this substation is the starting point of the CLSGinterconnection to Sierra Leone and Liberia. It should be noted that Linsan is amajor crossroad in Guinea with many interconnection lines. The creation of asecond substation close to Linsan with a direct connection between the twoshould thus be looked at. This will increase the safety and the security of thenetwork.

The substations of Sikasso is a good connection point in Mali due to its highpotential for renewable energy and the existing interconnections to neighboringcountries. This line will thus connect to the Mali-Burkina-Ghana 330 kV line whichshould be commissioned by 2022.

12 Les valeurs présentées sont soumises à l'hypothèse initiale de flux et de production et sont données ici commemoyen d'identifier les avantages de la nouvelle interconnexion

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 93/176

Fina

lver

sion

The total distance of this line from Tobene, Linsan and Sikasso is of approximately1600 km. Due to the long distances between these potential substations,additional substations should be created in order to increase the transmissioncapabilities of the line and allow for an easier operation of this one. For the sakeof the simulations performed, intermediate substations were modelled at Soma,in The Gambia, and at Mansoa in Guinea Bissau. These intermediary substationsand the precise path of the line should be defined based on detailed specificenvironmental and technical studies. The transfer capacity of this 330kVbackbone should be further increased thanks to these intermediate substationscoupled with an adequate compensation scheme.

Figure 47: Proposed path of the 330 kV Western backbone

It is noted that the construction of this new 330 kV double circuit westernbackbone avoids the construction of several other sections along its path. Thesesections include the Linsan – Manantali – Bamako – Sikasso lines as well as the225 kV Senagal lines connecting Tobene to the OMVG line as well as sections ofits eastern part (Kaolack-Tambacounda-Kedougou-Mali-Labé-Linsan). It wasseen that considering the load levels for the study period and the plannedexchanges, these lines are overloaded in N-1 conditions if the western backboneis not constructed.

An AC solution is here preferred compared to a DC one for two main reasons.First, the region being still slightly meshed, this new AC corridor will allow toincrease the grid stability and reinforce the synchronization of the neighboringcountries together. Secondly, operating an HVDC line in parallel with AC line is anew challenge for the region and necessits some specific actions to beimplemented to support a contingency in the area. Furthermore, creatingintermediate HVDC substations is more difficult and costly.

CONNECTION OF WESTERN AND MEDIAN BACKBONES

New 330 kV double circuit line Bobo – Ferkéssedougou. This new line will createa direct link between both the Western Backbone and the Median backbone inorder to create a single 330 kV double circuit link running from Senegal to Nigeria.The new line will follow the path of the existing 225 kV line from Bobo toFerkéssedougou.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 94/176

Fina

lver

sion

SECOND LINE OF THE COASTAL BACKBONE (NIGERIA – GHANA)

As seen in the conclusions of the dynamic studies realized for 2022, this projectis of great importance for the synchronization of Nigeria to the rest of the WAPPand to allow for a stable operation of the network under contingency andconsideration of the large economic exchanges planned. Additionally to the levelof imports and exports planned from Nigeria, the location of the new CCGT unitat Maria Gleta, connected on this line justifies the second line in order to allow asecure evacuation of the power in case of contingency. Furthermore, during timeswhen countries from the center of the WAPP (Côte d’Ivoire, Mali, Ghana, BurkinaFaso) are exporting a lot of power to Nigeria, the loss of the Ghana – Nigeriacoastal single circuit is not sufficient to securely operate and overloads are seenon the 161 kV grid of Togo and Benin.

The exact path of the line as well as the substations between Benin and Nigeriaare still to be determined. The connection of this line to Sakete and to the Lagosregion seems the most reasonable at this time considering the load levelsexpected. Due to environmental and technical feasibility constraints, a new pathmay be necessary. It is here noted that the substation of Onigbolo has beendetermined as a potential substation by the WAPP.

NEW 330 KV DOUBLE CIRCUIT LINE SALKADAMNA – KATSINA

This new 330 kV double circuit line connects in the existing 330 kV substation ofSalkadamna (Niger), to a new 330 kV substation in Malbaza (Niger) and Gazoua(Niger) and to the existing 330 kV substation in Katsina (Nigeria). The total lengthof the line is estimated to be of around 500 km. This line is needed in the longterm horizon in order to allow for the increase of load in the NCE (Niger Centre-Est) region of Niger. Due to the long distances currently connected through 132kV lines, the voltage drops in the region are significant and this new voltage levelwill hold the voltage in the operating range in this region.

Furthermore, at the 2033 horizon, this line will allow to export the solar power fromthe Northern region of Niger to Nigeria.

NEW 225 KV LINE SAN PEDRO – TIBOTO - BUCHANAN

This interconnection line between Liberia and Côte d’Ivoire along the coast isexpected to be commissioned in 2026 at the same time as the Tiboto power plantand goes hand in hand with this hydro project. The line will connect at San Pedro225 kV substation in Côte d’Ivoire and at Buchanan in Liberia.

NEW 225 KV LINE TENGRELA – SYAMA - BOUGOUNI

Mali and Côte d’Ivoire both have the intention of connecting mines to theirinterconnected network. These mines are located in the North of Côte d’Ivoirearound Tengrela and in the South of Mali around Syama. Due to the short distanceof about 40 km between these two sites, it is clear that there is a regional interestin connecting both of the sites together and thus creating a new interconnectionbetween Mali and Côte d’Ivoire. This new single circuit line will run from Bougouni(Mali) to Syama (Mali) and Tengrela. This line will allow to satisfy the security ofsupply of these mines by respecting the N-1 rules and will increase the systemstability in the case of the loss of the interconnection between Sikasso andFerkéssedougou.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 95/176

Fina

lver

sion

NEW 330 KV DOUBLE CIRCUIT LINE BOLGATANGA – JUALE – DAWA

This 330 kV double circuit line connecting the North and the South of Ghana hasbeen defined as a WAPP priority project. This project is planned from Bolgatangain the north of Ghana and to Juale and Dawa to the south. In order to double theBolgatanga – Tamale corridor, it is proposed that this line drops down to connectat the 330 kV substation in Tamale and in Yendi to connect with the medianbackbone and avoid a possible bottleneck on the Yendi – Juale 161 kV line. Thisline offers the advantage of doubling the North-South corridor in Ghana whichproves to be a necessity considering the N-1 security criteria. Furthermore, thisline creates a more direct path to conduct the renewable energy from Burkina andthe North of Ghana to the South of Ghana near Accra, where a large part of theload is located.

Figure 48: New 330 kV line Bolgatanga - Juale – Dawa

EASTERN BACKBONE

This 330 kV double circuit line has as objective to connect the Northern region ofNigeria to the Southest region. The line which has an estimated length of 1856km will connect at the substations of Calabar, Ikom, Ogoja, Kashimbilla, Mambilla,Jalingo, Yola, Hong, Biu, Damaturu, Potiskum, Azare, Dutse, Jogana as well asa section from Sokoto to Kaura and Katsina. This project is in line with the TCNmaster plan and will allow for:

· A smooth integration of renewable energy (hydro (Mambilla), solar and wind)· A significant increase of the load in all regions of Nigeria· An increase of the security of supply in Nigeria· An increase in the exchanges in the WAPP region

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 96/176

Fina

lver

sion

Additionnaly, this line will be of necessity in the case of a connection of the WAPPto the Central African Power Pool (CAPP).

REINFORCEMENT OF OMVG LOOP WEST

At the 2033 horizon, Guinea Bissau and The Gambia are both expected to beimporting a large part of their power from Guinea and Senegal. In the short term,both of these countries are planned to be interconnected through the single circuitOMVG loop which allows for a thermal capacity of around maximum 330 MVA.Considering the imports of The Gambia and Guinea Bissau in 2033 at the peakevening time which account for up to around 350 MW, it is clear that the singlecircuit becomes insufficient to respect the N-1 criteria with such levels of import.In order to satisfy the security of supply in these countries, a second line isplanned on the western part of the OMVG loop connecting Kaolack to Kaleta.Depending on the possibilities and specific studies to be undertaken, thepossibility of connecting the second line of double circuit directly through Kaolack– Brikama – Soma - Tanaff – Mansoa – Bambadinca – Saltinho and Kaleta ispresented in the following figure.

Figure 49: Proposed path of the second circuit of OMVG West

Up to 2025, due to the importing nature of Senegal before the apparition ofcombined cycles, the OMVG single circuit line is loaded to levels which do notsupport the N-1 criteria. The most critical sections of this line concern the partsbetween Linsan and Guinea Bissau which are loaded to higher levels due to theimports of Senegal, The Gambia and Guinea Bissau.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 97/176

Fina

lver

sion

3.4.2. Modelling of the 2033 WAPP networkThe hypothesis and information that was used to create the 2033 model of theWAPP are described in the following sections.

3.4.2.1. NATIONAL REINFOCEMENTS

The list of national reinforcements that were done on the high voltage grid for eachcountry in 2033 is described in Annex.


The load levels which were modelled in the different scenarios that were studiedfor 2033 are details in the following paragraph.

For the target year of 2033, three different scenarios were analyzed:

· Asynchronous peak evening situation· Peak renewable scenario· Synchronous Off-peak scenario

The load levels modelled in each of these scenarios is presented here under.

In the peak load scenario, the load modelled corresponds to the yearly peak loadof every country. This scenario where every country observes their peak load atthe same time is a conservative way of analyzing the reinforcement needs on thegrid. This active load level is presented Table 25. The same power factor as theexisting model was kept for this study year.

Country Peak Load2033

BENIN 704 MW

BURKINA 1043 MW

CIV 3981 MW

GAMBIE 297 MW

GHANA 4957 MW

GUINEE 1104 MW


LIBERIA 411 MW

MALI 1118 MW

NIGER 1063 MW

NIGERIA 20850 MW13

SENEGAL 2065 MW

13 As the demand forecast for Nigeria was based on the TCN Master Plan, the same load level was modelled as inthis reference. This allows for a close accordance of the reinforcement needs between both studies.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 98/176

Fina

lver

sion

Country Peak Load2033

SIERRA LEONE 696 MW

TOGO 646 MW

TOTAL 39151 MW

Table 25: Load level – Peak 2033

The objective of the renewable scenario is to evaluate if the grid is sufficientlymeshed to evacuate all renewable power while respecting the N-1 criteria. Forthis matter, the average daily solar peak of 13h is chosen to represent thissituation. The demand level of each country in this scenario is shown in Table 26.

Country % of PeakLoad

Renewable ScenarioLoad

BENIN 67 % 476 MW

BURKINA 59 % 616 MW

CIV 71 % 2834 MW

GAMBIE 82 % 244 MW

GHANA 67 % 3330 MW

GUINEE 82 % 906 MW

GUINEE BISSAU 82 % 177 MW

LIBERIA 82 % 337 MW

MALI 54 % 602 MW

NIGER 38 % 399 MW

NIGERIA 85 % 17787 MW

SENEGAL 66 % 1371 MW

SIERRA LEONE 82 % 571 MW

TOGO 67 % 436 MW

TOTAL 77 % 30087 MW

Table 26: Load level – Renewable scenario 2033

In this scenario, the power generation from PV plants reaches a level of more than60% of the power generation. Such a level of renewable penetration may causesome problems for the system stability. These problems which are described inmore details in section 3.4.4 are not captured in the static analysis which areperformed for the target year of 2033. A dedicated study is required to define theoperational constraints and measures to be taken to allow this important share.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 99/176

Fina

lver

sion

The off-peak scenario was analyzed for the 2033 network which was developed.Based on the average load curve of the WAPP, this synchronous off-peaksituation appears around 9am. Based on this information, the followingassumptions were made concerning the production dispatch and the load. Theload level modelled in this scenario is presented in Table 27.

Country % of PeakLoad

Off-peak Load

BENIN 46 % 324 MW

BURKINA 57 % 595 MW

CIV 52 % 2070 MW

GAMBIE 44 % 131 MW

GHANA 67 % 3321 MW

GUINEE 44 % 486 MW

GUINEE BISSAU 44 % 95 MW

LIBERIA 44 % 181 MW

MALI 61 % 682 MW

NIGER 53 % 563 MW

NIGERIA 52 % 10842 MW

SENEGAL 58 % 1198 MW

SIERRA LEONE 44 % 306 MW

TOGO 46 % 297 MW

TOTAL 51 % 21091 MW

Table 27: Load level – Off-peak 2033

3.4.3. Static StudiesThe results of the static studies realized for the target year of 2033 for the differentscenarios are shown in the following paragraphs. The methodology andassumptions taken for those scenarios were described in the previous section.

3.4.3.1. ASYNCHRONOUS PEAK 2033

Similarly as what was done for 2022 and 2025, the asynchronous peak, in whichevery country experiences its peak at the same time, has been modelled in thepeak scenario. The country balances resulting from the economic exchanges ispresented in Table 28.

Country Peak Balance

BURKINA - 901 MW

CIV - 13 MW

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 100/176

Fina

lver

sion

GAMBIE - 148 MW

GHANA 39 MW

GUINEE 1287 MW


LIBERIA - 145 MW

MALI - 516 MW

NIGER - 540 MW

NIGERIA 1528 MW

SENEGAL 119 MW



Table 28: Country balance – Peak 2033

It is seen that in the evening peak of this target year of 2033, Burkina Faso isimporting up to almost 90% of its load. The other importing countries are mainlyMali and Niger. On the other hand, Nigeria is exporting more than 1500 MWthanks to the gas availabilities and the construction of CCGTs and Guinea isexporting a large part of their hydro generation which is clearly in excesscompared to their national load.

The flows on the interconnection lines are shown here in Table 29.


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)






Kara_330-Yendi_330-1 330 TB GH 146.1 19.9

Kara_330-Yendi_330-2 330 TB GH 146.1 19.9

Linsan_330-Sikasso_330-1 330 GU MA 145.2 25.2

Linsan_330-Sikasso_330-2 330 GU MA 145.2 25.2








WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 101/176

Fina

lver

sion


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)





Tiboto _225-Buchanan_225-1 225 CI LI 96.6 30.7

Mali_225-Sambangalou_225-1 225 GU SE 81 25.7



New Agbara_330-Sakete_330-1 330 NI TB 71.7 27.2




Davié_330-Dawa_330-1 330 TB GH 63.5 6.4

Davié_330-Dawa_330-2 330 TB GH 63.5 6.4

Davié_330-Dawa_330-3 330 TB GH 63.5 6.4





Tamale_330-Ferkessedougou_330-1 330 GH CI 51.7 11.8



Tobene_330-Linsan_330-1 330 SE GU 39.9 23.3

Tobene_330-Linsan_330-2 330 SE GU 39.9 23.3




Kaolack_225-Soma_225-1 225 SE GA 36.3 14.2

Ferkessedougou_330-Bobo_330-1 330 CI BU 34 5.4

Ferkessedougou_330-Bobo_330-2 330 CI BU 34 5.4





Dunkwa_330-Bingerville_330-1 330 GH CI 27.2 5.4

Kaolack_225-Brikama_225-1 225 SE GA 27 9.7

Kaolack_225-Brikama_225-2 225 SE GA 27 9.7

Man_225-Yekepa_225-1 225 CI LI 25.8 7.6

Man_225-Yekepa_225-2 225 CI LI 25.8 7.6

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 102/176

Fina

lver

sion


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)

Tengrela_225-Syama_225-1 225 CI MA 23.5 7.5

Tanaf_225-Soma_225-1 225 SE GA 19.3 5.9

Tanaf_225-Soma_225-2 225 SE GA 19.3 5.9

Tanaf_225-Soma_225-3 225 SE GA 19.3 5.9

Mano_225-Kenema_225-1 225 LI SL 13.6 4.5

Mano_225-Kenema_225-2 225 LI SL 13.6 4.5

N'Zérékore_225-Yekepa_225-1 225 GU LI 13 7.5

N'Zérékore_225-Yekepa_225-2 225 GU LI 13 7.5

Katsina_132-Gazaou_132-1 132 NI NR 13 13.8

Sikasso_225-Ferkéssédougou_225-1 225 MA CI 11.9 8.6

Cinkassé_161-Bawku_161-1 161 TB GH 9 6.9

Bakel_225-Kayes_225-1 225 SE MA 7.2 9.1

Zabori_330-Malanville_330-1 330 NR TB 3 0.7




Table 29: Flows on interconnection - Peak 2033

In 2033, the security analysis has shown that no contingency is problematic onthe regional high voltage grid. The list below shows the contingencies which cancause a problem on the national grid. These contingencies do not have a regionalimpact and should be treated by the national master plans of the countryimpacted.


Sikasso 225 kV - Koutiala 225 kV-1 (NATIONAL) MA Voltage Collapse in Mopti

Toulepleu 225 kV – Zagne 225 kV – 1 (NATIONAL) CI Voltage Collapse in the area of Mine Ity

Dueckoue 225 kV – Zagne 225 kV – 1 (NATIONAL) CI Voltage Collapse in the area of Mine Ity

Kounoune 225 kV – Cap des Biches 225 kV (NATIONAL) SEN Overload of second circuit (>110%)

Mboro 225 kV – Tobene 225 kV (NATIONAL) SEN Overload of second circuit (>110%)

Table 30: List of problematic contingencies (NATIONAL) – Peak 2033

3.4.3.2. RENEWABLE INTEGRATION SCENARIO

The results of the scenario in which a maximum renewable injection on the gridwas assumed is shown in this section. Based on the results of the economicanalysis, the balances of each country in this scenario is shown in Table 31. Itcan be seen that this scenario assumes that Nigeria is importing a lot ofgeneration: up to 2500 MW.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 103/176

Fina

lver

sion

Country Renewable scenario - Balance

BURKINA 283 MW

CIV 300 MW

GAMBIE - 37 MW

GHANA 1205 MW

GUINEE -22 MW


LIBERIA - 123 MW

MALI 433 MW

NIGER 1007 MW

NIGERIA - 2515 MW

SENEGAL - 242 MW



Table 31: Country balance – Renewable Scenario 2033

It is seen that in this scenario, the situation of Burkina and Nigeria have beencompletely inversed compared to the evening peak situation. This can beexplained by the renewable potential of the different countries and their loadlevels. Furthermore, the exporting countries are now Niger, Ghana, Guinea andCôte d’Ivoire.

In this scenario, renewable infeed from wind turbines is set to 100 % and that fromPV parks is set to 63 % which correspond to a typical maximum production froma PV park in West Africa. For irrigation purposes, the hydro production was set to10% of the maximum producible. The rest of the generation is produced bythermal units such as must run units and most economic combined cycle units.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 104/176

Fina

lver

sion

Generation type Generated Active Power (MW)

Hydro 1230 MW

Solar 20037 MW

Wind 1700 MW

Thermal 8456 MW

Table 32: Generation dispatch – Renewable Scenario 2033

As mentioned in the modelling section for the 2033 target year, such level ofrenewable penetration as presented in this renewable scenario should be thepoint of dedicated studies to assess the feasibility of operating stably under theseconditions.

Table 33 shows the flows on each interconnection line in the renewable scenariofor the 2033 target year.


Country -Sending

Node

Country –Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)



Parakou_330-Kainji_330-1 330 TB NI 300.8 37.8

Parakou_330-Kainji_330-2 330 TB NI 300.8 37.8


Yendi_330-Kara_330-1 330 GH TB 290.7 36.0

Yendi_330-Kara_330-2 330 GH TB 290.7 36.0

Dawa_330-Davié_330-1 330 GH TB 235.4 25.7

Dawa_330-Davié_330-2 330 GH TB 235.4 25.7

Dawa_330-Davié_330-3 330 GH TB 235.4 25.7

Sakete_330-New Agbara_330-1 330 TB NI 225.7 32.9

Sakete_330-New Agbara_330-2 330 TB NI 225.7 32.9

Gazaou_330-Katsina_330-1 330 NR NI 217.3 27.9


Sikasso_330-Linsan_330-1 330 MA GU 119 26.2

Sikasso_330-Linsan_330-2 330 MA GU 119 26.2

Ferkessedougou_330-Tamale_330-1 330 CI GH 102.1 19.9

Ferkessedougou_330-Tamale_330-2 330 CI GH 102.1 19.9

Ouaga Est_330-Goroubanda_330-1 330 BU NR 84.5 20.3

Ouaga Est_330-Goroubanda_330-2 330 BU NR 84.5 20.3

Kayes_225-Bakel_225-1 225 MA SE 83.9 42.3





WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 105/176

Fina

lver

sion


Country -Sending

Node


Node

ActivePower Flow

(MW)Loading -

Current (%)



Manantali_225-Boureya_225-1 225 MA GU 64.6 23.1

Manantali_225-Boureya_225-2 225 MA GU 64.6 23.1





Linsan_330-Tobene_330-1 330 GU SE 55.9 17.4






Sanakoroba_225-Siguiri_225-1 225 MA GU 31.2 15.7

Sanakoroba_225-Siguiri_225-2 225 MA GU 31.2 15.7

Mano_225-Kenema_225-1 225 LI SL 30.5 10.8

Mano_225-Kenema_225-2 225 LI SL 30.5 10.8

Dosso_132-Birnin Kebbi_132-1 132 NR NI 30.1 65.9

Boundia_225-Morisanako_225-1 225 CI GU 27.5 10.4

Boundia_225-Morisanako_225-2 225 CI GU 27.5 10.4






Aflao Ghana_161-Lomé (Aflao) 1_161-1 161 GH TB 18 17.8


Tanaf_225-Soma_225-1 225 SE GA 17.4 5.9

Tanaf_225-Soma_225-2 225 SE GA 17.4 5.9

Tanaf_225-Soma_225-3 225 SE GA 17.4 5.9








Man_225-Yekepa_225-1 225 CI LI 10.1 4.2

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 106/176

Fina

lver

sion


Country -Sending

Node


Node

ActivePower Flow

(MW)Loading -

Current (%)

Man_225-Yekepa_225-2 225 CI LI 10.1 4.2



Ferkessedougou_330-Bobo_330-1 330 CI BU 2.9 1.8


Brikama_225-Kaolack_225-1 225 GA SE 2 10.0

Brikama_225-Kaolack_225-2 225 GA SE 2 10.0

Table 33: Flows on interconnection - Renewable Scenario 2033

No regional security problems were detected when analysis this renewablescenario. The additional national contingencies recorded compared to the peakscenario are defined in Table 34.

Contingency Country Overload Mitigation measure

Akoupe Zeudji – Abobo 225 kV

Yopougon 3 – Azito 225 kV

Azito – Vridi 225 kV

Abobo – Djibi 225 kV

CI

The loss of one of these lines gives anoverload on one of the others.

The reason for this is that nogeneration is dispatched in Abidjan.

Turn on some thermalgeneration in Abidjan inorder to remove theoverload.

Table 34: List of problematic contingencies (NATIONAL) – Renewable scenario 2033

3.4.3.3. SYNCHRONOUS OFF-PEAK 2033

The generation dispatch considered is similar to the one presented in therenewable scenario. Production from solar PV parks was however put to 50 % ofthe production of the renewable case. Generation from hydro power plants wasset to 10% for irrigation purposes.

Generation type Generated Active Power (MW)

Hydro 1230 MW

Solar 10019 MW

Wind 1700 MW

Thermal 8923 MW

Table 35: Generation dispatch – Off-peak 2033

Based on these assumptions reflecting the results of the economic analysis andthe daily load curves, the country balances in the 2033 off peak situation is shownin the following table.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 107/176

Fina

lver

sion

Country Off-peak scenario - Balance

BURKINA -114 MW

CIV -21 MW

GAMBIE - 28 MW

GHANA -124 MW

GUINEE 225 MW


LIBERIA - 61 MW

MALI - 50 MW

NIGER 322 MW

NIGERIA 527 MW

SENEGAL - 419 MW



Table 36: Country balance – Off-peak Scenario 2033

The flows on the lines is represented in the table below.


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)







Kara_330-Yendi_330-1 330 TB GH 108.8 14.8

Kara_330-Yendi_330-2 330 TB GH 108.8 14.8

Davié_330-Dawa_330-1 330 TB GH 105.8 11.6

Davié_330-Dawa_330-2 330 TB GH 105.8 11.6

Davié_330-Dawa_330-3 330 TB GH 105.8 11.6

Sikasso_330-Linsan_330-1 330 MA GU 103.1 26.7

Sikasso_330-Linsan_330-2 330 MA GU 103.1 26.7






WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 108/176

Fina

lver

sion


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)

Kayes_225-Bakel_225-1 225 MA SE 74.3 40.3



Bolgatanga_330-Bobo_330-1 330 GH BU 69 9.9

Bolgatanga_330-Bobo_330-2 330 GH BU 69 9.9

Lomé (Aflao) 1_161-Aflao Ghana_161-1 161 TB GH 55.6 47.2













Tanaf_225-Soma_225-1 225 SE GA 35.2 11.7

Tanaf_225-Soma_225-2 225 SE GA 35.2 11.7

Tanaf_225-Soma_225-3 225 SE GA 35.2 11.7

Bolgatanga_225-Ouaga Sud_225-1 225 GH BU 34 17.7



Boundia_225-Morisanako_225-1 225 CI GU 30 13.6

Boundia_225-Morisanako_225-2 225 CI GU 30 13.6


Man_225-Yekepa_225-1 225 CI LI 28.5 8.5

Man_225-Yekepa_225-2 225 CI LI 28.5 8.5





Mano_225-Kenema_225-1 225 LI SL 23.7 11.5

Mano_225-Kenema_225-2 225 LI SL 23.7 11.5





WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 109/176

Fina

lver

sion


Country -Sending

Node

Country -Receiving

Node

ActivePower Flow

(MW)Loading -

Current (%)

Zabori_330-Malanville_330-1 330 NR TB 16.3 2.8

Brikama_225-Kaolack_225-1 225 GA SE 13.9 7.3

Brikama_225-Kaolack_225-2 225 GA SE 13.9 7.3




Kodeni_225-Ferkéssédougou_225-1 225 BU CI 10 9.7








Table 37: Flows on interconnection Off-peak Scenario 2033

In these simulations, it was observed that all voltages were maintained in thecorrect operating range. However, it can be seen that many generators areabsorbing reactive power during this off-peak scenario. Specific dynamic analysisshould be carried out to determine whether this situation cause stability issues.As the WAPP network is made of an increasing number of high voltage lines,specific reactive power compensation studies should be done to determine theamount of inductances required for the commissioning of each new high voltageline.


Yopougon 3 225 kV – Songon 225 kV (NATIONAL) CI Overload of second circuit when thepower of Songon is at its maximum

Table 38: List of problematic contingencies (NATIONAL) – Off peak 2033

3.4.4. Operating the network with an increasing share ofrenewablesWith the new capacity of renewable power generation which is planned to connectto the grid in the 2033 horizon, new operational limits can be reached.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 110/176

Fina

lver

sion

Operating a power system with high instantaneous penetration of renewablesposes numerous operational challenges which must be addressed to ensure ahigh system reliability. These operational challenges can be classified dependingon the timescale of interest. A short non-exhaustive overview is given below.These different aspects should be studied in a dedicated study to define theoperational constraints and measures to be taken to allow the important share ofrenewable. These studies should comprise of a review of the grid code, such asthe WAPP Operational Handbook, as well as a renewable integration study whichwill study the points mentioned in the following paragraphs.

Within the timescale of minutes to hours, both the variability and especially thelimited predictability of renewables (PV and wind power) may have a direct impacton the required operating reserves guaranteeing the balance between generationand demand.

First, the variability of such units adds to the overall power imbalance that drivesthe amount and flexibility of the reserves. Forecast errors on the other hand createan uncertainty and require real-time relatively higher load following or rampingreserves.

Secondly, especially with a very high penetration, such units will replacetraditional power plants, which normally guarantee the provision of the requiredreserves and ancillary services. Additional measures have therefore to be takento guarantee the balance between generation and demand at all times.

Operational impacts at faster timescales (seconds to minutes) are mainly relatedto the way these units are interfaced with the network. The strong coupling of atraditional synchronous generator to the grid ensures a high short-circuit poweras well as an inertial response. On the contrary, converters typically transformdirect current (DC) electricity to AC power by controlling semiconductor devicesat a high switching frequency. Due to this intermediate DC link which decouplesthe generator from the grid, no inertial response is provided, and the converterinterface also inherently weakens or eliminates the response to grid faults.

Instead of the physical characteristics of the synchronous machine, the controlstrategy of the converter predominately determines the electrical dynamicinteraction with the system. However, both the high short-circuit power as well asthe inertial response are essential for the operation of current transmission grids.A high and sustained short-circuit current during grid faults, will limit the impactarea of voltage dips and triggering fault detection by protection relays. The inertialresponse on the other hand will limit the rate of change of frequency (ROCOF)after a power imbalance in the system and as such inherently provides time forthe governors and turbines to react in order to stabilize the system frequency.Less inertia immediately leads to higher ROCOF values and lower minimumfrequencies for the same considered incident as given in the figure below. Withouttaking additional measures, this will lead to large scale load shedding or trippingof protection relays.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 111/176

Fina

lver

sion

Figure 50: Impact of inertia on the rate of change of frequency (ROCOF) [Entso-e]

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 112/176

Fina

lver

sion

4. OPPORTUNITIES BEYOND THE BORDERSOF THE WAPP

4.1. Possible interconnection with North Africa

In parallel to the main study, the consultant was asked to carry out a pre-feasibilitystudy on an interconnection between North-Africa and WAPP. This study includesa technical and an economic part.

The objective of the technical study is to study the technical feasibility of aninterconnection between the North-African power grid and the WAPP electricitygrid. In particular, it analyzes the impact of technology (AC or DC) on stability ofthe system as a whole.

Based on the results and recommendations from the technical analysis, aneconomic study is carried out. The objective of this study is to establish, on theone hand, the optimal exchanges on interconnection, and on the other hand thebenefits for the West African system of an interconnection with North-Africa.

4.1.1. Technical study

4.1.1.1. INTRODUCTION

In this study, the prefeasibility of the connection between the North-African powersystem and the WAPP-system is further analysed. This part of the study focuseson the technical aspects, and more specifically on the way the system stability isinfluenced when an AC or DC line is used to link both systems.

The goal is mainly to address the question whether it is possible to use ACtechnology for the connection, taking into account the long distances andassumed power transfers between the two systems.

Secondly, also the options to use DC technology are shortly addressed and themain advantages and disadvantages of using LCC or VSC are presented.

Finally, the congestions within the Moroccan power system are investigatedconsidering 1000 MW export produced by renewable energy sources (e.g. PV)close to the 400kV bus LAAYOUNEII.

4.1.1.2. MODELLING

4.1.1.2.1. Morocco and WAPP-systemThe WAPP model has been developed in the power system simulation toolEurostag. The power system of Morocco is provided in PSS/E. Therefore, aconversion from the static and dynamic model of Morocco is firstly been done tomerge the two systems in Eurostag. As a starting point, each system is balancedby itself such that no large power transfers will take place in case the systems arelinked by an AC line.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 113/176

Fina

lver

sion

4.1.1.2.2. European system equivalentIn this prefeasibility study, the European power system is represented by anequivalent model which will mimic the behaviour of the European network and itsdynamic interactions with the Moroccan system. The applied modelling detail isconsidered to be sufficiently accurate for the type of studies presented in thisreport.

The equivalent model is connected to the substation PINARREY (Spain). All otherlines linking this bus with the remaining Spanish system are being disconnected.Only the lines with TARIFA (1&2) remain in operation in the study. See also Figure51 below.

Figure 51: European system equivalent and surrounding network

For the static simulations, such as the AC load flow, an equivalent generatorlinked with a transformer and an active power setting equal to the net transferfrom PINARREY to TARIFA (within the fully detailed system model this is setequal to 0 MW) is applied.

For the dynamic simulations, the generator is equipped with a governor (IEEEG1)and a voltage regulator (IEEET1). The parameters for these regulators andgenerator as well as the type of model used are based on the consultantexperience and are shown in the tables below. These parameters are the oneusually applied in system studies. Due to the type of study requested, the mostimportant parameters are the inertia, the equivalent size of the unit and the Kparameter of the model IEEEG1.

Parameter Value Parameter Value

SN (MVA) 360 000 H 5

Pmax MW) 306 000 D 0,0

Pmin (MW) 0 Xd 2.3

Qmax (MVar) 190 000 Xq 2

Qmin (MVar) -190 000 X’d 0,29

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 114/176

Fina

lver

sion


Xsource 0,2 X’q 0,3

T’d0 6,5 X’’d=X’’q 0,2

T’’d0 0.03 Xl 0,15

T’q0 0,5000 S(1.0) 0,2

T’’q0 0,15 S(1.2) 0,4

Table 39: Generator parameters for the ENTSO-E (European) system equivalent


K 2,2400 K2 0,0000

T1 0,0000 T5 0,0000

T2 0,0000 K3 0,0000

T3 (> 0) 0,5000 K4 0,0000

Uo 0,1000 T6 0,0000

Uc (< 0) -0,1000 K5 0,0000

PMAX 1,0000 K6 0,0000

PMIN 0,0000 T7 0,0000

T4 0,5000 K7 0,0000

K1 1,0000 K8 0,0000

Table 40: Governor (IEEEG1) parameters for the ENTSO-E (European) system equivalent


TR (sec) 0 KF 0,2

KA 400 TF (>0)(sec) 1

TA 0,02 Switch 0,0000

VRMAX 6,02 E1 2,8875

VRMIN -6,02 SE(E1) 0,39

KE 1 E2 3,85

TE (>0)(sec) 0,015 SE(E2) 0,5630

Table 41: Exciter (IEEET1) parameters for the ENTSO-E (European) system equivalent

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 115/176

Fina

lver

sion

4.1.1.2.3. InterconnectionThe connection will link the Moroccan/European system and the WAPP-system,more specifically, the south of Morocco and Senegal with an intermediate tappingto Mauritania. The minimum distance and optimal power evacuation capability ofthe surrounding network are the main criteria for determining the differentconnection points and routing of the line.

In Morocco, it is chosen to use the substation named DAKHLA as starting pointof the interconnection as it is one of the most southerly located 400 kV substationsin the country. It is also directly linked with the long 400 kV (double circuit)backbone going from south to north Morocco.

In Senegal different connection options do exist, but in this study the substationTOBENE is chosen as it is linked to the 225 kV grid of the country. Furthermore,as it is one of the largest substations in the country, it can be assumed to becomepart of any 330/400 kV network upgrade, linking Senegal and the rest of theWAPP-system.

The intermediate tapping in Mauritania is located at the Nouakchott substationwhich is part of the 225 kV network. The line length DAKHLA-NOUACKCHOTT isestimated to be equal to 800 km, the length between NOUACKCHOTT-TOBENEis taken equal to 450 km. Figure 52 gives an overview of the link and theconnection points.

Figure 52: Proposed 400 kV AC interconnection linking the existing system (new equipment is shown in black, theexisting busses are coloured in blue)

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 116/176

Fina

lver

sion

4.1.1.3. ANALYSIS OF THE AC OPTION

4.1.1.3.1. Description and approachA double-circuit overhead line operated at 400 kV is applied with a maximumcapacity of 1000 MVA each. As such, in an N-1 situation, still 1000 MVA cantheoretically be transferred over the interconnection. To connect the line with the225 kV substation in TOBENE, three transformers of 500 MVA each are used(see also Figure 52). The same applies for the connection with theNOUACKCHOTT substation. For the lines and transformers, standardparameters are being used as given in the table below.


R 0,033 Ω/km Length DAKHLA-NOUACKCHOTT 800 km

X 0,33 Ω/km Length NOUACKCHOTT-TOBENE 450 km

B/2 1,54 µS/km

Table 42 : Parameters of 400 kV line

When operating such a long AC overhead line, several (dynamic stability) issuesmay occur. A non-exhaustive list is presented below:

· Due to the Ferranti effect, reactive power compensation (e.g. in the form ofshunt reactors) need to be installed to prevent overvoltages on the line duringenergization.

· During loading, voltage control may become an issue: with such long AC lines,compensation is often needed, which can give rise to new problems, such assubsynchronous resonance (i.e. the resonant frequency of turbine generatorshafts coincides with a resonant frequency of the system.)

· Fault protection issues: timing problems due to communication delays betweenthe different parts of the protection equipment (measurement devices, circuitbreakers, …) may lead to untimely fault clearance or maloperation of theprotection system.

· Transient stability issues may occur. Especially during high loading of the line,generators can lose synchronism after a fault occurs on the interconnectiondue to the low synchronizing power between the two systems(Morocco/Europe & WAPP). Even in case a fast fault clearance is ensured, asystem split is difficult to prevent and additional measures have to be taken toenhance the transient stability.

In this report, the latter form of stability is investigated in more detail. To this end,a short-circuit is applied on one of the lines between DAKHLA andNOUAKCHOTT (95% of the line length starting from DAKHLA) for differentloading levels of the line. The fault is cleared after 100 ms by opening both endsof the line. It is investigated whether the system is able to remain in synchronism.The line loading is increased by installing a (dummy) generator and load at bothends of the line and gradually increase the active power setting of both.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 117/176

Fina

lver

sion

In the model, shunt reactors are installed at the three busses to compensate forthe generated reactive power during low loading of the line. A simple approach tocalculate the required size is applied, i.e. the reactors fully compensate thesusceptance of the line. More elaborate ways to compensate the line duringdifferent loading levels using switched reactors/, series compensations or SVC’sare possible but are out of the scope of this study.

4.1.1.3.2. Simulation resultsFirstly, the transient stability is investigated for (almost) zero active power transferover the lines. Nevertheless, it became clear that, even for such low transfer, thegenerators closely located to NOUAKCHOTT lose synchronism after the fault wascleared.

This is clearly shown in Figure 53 and Figure 54 (simulation results may assumedto be valid until system loses synchronism, thereafter protection systems aretriggered, which are not included in the EUROSTAG model). In these figures, theangles of some generators spread over the system and the power transfers(active and reactive) over the AC interconnection are given. All angles areexpressed with respect to the centre of inertia (since Europe has such high inertia,this centre lies closely to the border of Spain). Although most of the generatorsstay synchronized, the machines inside Mauritania go out-of-step.

In case no connection to NOUAKCHOTT is made (by disconnecting thetransformers), a higher active power transfer is feasible without jeopardizing thetransient stability considering the assumed fault location/duration, see also Figure55 and Figure 56 for an active power transfer of 200 MW.

This simulation is repeated to identify the maximum power transfer at which thesystem is still capable of remaining synchronized. A value of 280 MW is found tobe the limit. Increasing the limit would be possible by a faster fault clearing or byimplementing different measures to increase the transient stability (fast-valving,SVC’s, more and stronger interconnections between the system,…). Goingbeyond that limit, a clear system split will occur after fault clearance. See forinstance Figure 57 and Figure 58 for a power transfer of 500 MW; the generatorangles inside the WAPP and Moroccan system completely diverge from eachother (simulation results may assumed to be valid until system loses synchronism,thereafter protection systems are triggered, which are not included in theEUROSTAG model).

Considering the simulation results and all the former listed issues that may occur,it can be concluded that, from a system stability point of view, it will be verychallenging to operate such a long AC link connected to two weak points in thenetwork. Additional investments will be required, such as several intermediatesubstations, voltage control devices, series compensation, …

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 118/176

Fina

lver

sion

Figure 53: Generator angles - Three-phase short circuit at 10s (clearance at 10.1s) - intermediate connection atNOUAKCHOTT – 0 MW power transfer

Figure 54: Active and reactive power flow over the AC interconnection - Three-phase short circuit at 10s (clearance at10.1s) - intermediate connection at NOUAKCHOTT – 0 MW power transfer

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 119/176

Fina

lver

sion

Figure 55: Generator angles - Three-phase short circuit at 10s (clearance at 10.1s) - No intermediate connection atNOUAKCHOTT – 200 MW power transfer

Figure 56: Active and reactive power flow over the AC interconnection - Three-phase short circuit at 10s (clearance at10.1s) – No intermediate connection at NOUAKCHOTT – 200 MW power transfer

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 120/176

Fina

lver

sion

Figure 57: Generator angles - Three-phase short circuit at 10s (clearance at 10.1s) - No intermediate connection atNOUAKCHOTT – 500 MW power transfer

Figure 58: Active and reactive power flow over the AC interconnection - Three-phase short circuit at 10s (clearance at10.1s) – No intermediate connection at NOUAKCHOTT – 500 MW power transfer

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 121/176

Fina

lver

sion

4.1.1.4. ANALYSIS OF THE DC OPTION

4.1.1.4.1. Overview of different options: technology and lay-outBesides considering an AC connection, it is interesting to look at the options oflinking both systems by using HVDC technology. HVDC offers many advantagesin this specific case compared to the long AC lines. It provides for instance adecoupling of both systems (system dynamics due to faults are not directlytransferred from one system to the other), enhanced controllability, lower linelosses (no skin effect and no reactive power transfer), … just to name a few.

Independently of the specific converter technology (VSC: voltage sourceconverter or CSC: current source converter), it is firstly important to look at themain lay-out of the DC connection.

Different lay-outs can be applied for the considered HVDC link, such as amonopolar or bipolar system with ground return or metallic return. A bipolarsystem with ground return is chosen as it offers increased redundancy and allowssome flexibility for future tappings/extensions. The increased redundancy isachieved because the system can still operate at 50% of its nominal rating in caseof an outage of one converter or line. Future tappings can consist out of a bipolarlayout or can be build using a single converter (monopolar tapping with groundreturn).

The proposed lay-out is presented in Figure 59. In this figure, an intermediatetapping is foreseen at NOUAKCHOTT. However, to reduce cost and controlcomplexity of the link, this tapping could be added in a later stage and a standardpoint-to-point connection would be constructed in a first step.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 122/176

Fina

lver

sion

Figure 59: Proposed bipolar HVDC link ±320 kV (new equipment is shown in black, the existing busses are coloured inblue, the connection between the converters and AC grid is not presented in detail)

With respect to the applied HVDC technology, one can chose between CSC andVSC converters. A list with the main differences of interest is presented in thetable below:

CSC VSC

Lower cost Higher cost

Lower converter losses Higher converter losses

Requires strong AC system Operates into weaker AC systems

DC network extension is not straightforward More flexibility in terms of DC networkextension (tappings/multi-terminal) andcontrol

Consumes always reactive power Flexible reactive power support + otherancillary services

Table 43: Comparison between CSC and VSC technology

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 123/176

Fina

lver

sion

Besides it additional cost and losses, VSC offers many advantages over CSC,especially for the investigated system. For CSC for instance, a relatively strongAC system at the point of common connection (PCC) is required, commonlyexpressed in terms of the short circuit ratio (SCR). In general, it is assumed thata SCR of 2-3 is required to ensure a proper operation of the CSC. For lower SCR,commutation failures may occur.

Taking into account the weak connection points (low short circuit power atDAKHLA: 1.459 GVA, NOUACKCHOTT:1.321 GVA TOBENE: 3.480 GVA),additional measures (adding synchronous condensers, STATCOMs, …) havetherefore to be taken to apply CSC technology for a rating of 1 GVA. VSC looksthus the most appropriate solution for the system especially when grid support isrequired or future extensions (and tappings) of the system are foreseen.

4.1.1.4.2. Grid support by HVDC: simulation resultsHVDC links not only decouple both AC systems such that faults in one system donot propagate to the other system, they can also provide grid support after anincident takes place. One of such support mechanismwhich would be of interestis to provide frequency control to the WAPP system. Since the Moroccan systemis linked with Europe, it has a substantial amount of inertia and primary reservewhich it can be offered to the WAPP system by modifying the power controllerwhich is demonstrated in the simulation results presented below.

A VSC HVDC connection is integrated in the system, connected betweenDAKHLA and TOBENE. In our power system simulation tool EUROSTAG,different standard control blocks are applied to model both VSC as CSC HVDCconnections. The layout itself (bipolar, monopolar, ground return of metallicreturn) of the point-to-point HVDC link doesn’t have that much influence on theway it is represented in the software. As long as we investigate standard forms ofpower system stability (voltage, angle, frequency, … stability) in which we aremainly interested in the dynamic interaction of the converters with the system, thesame model is used for different layouts. Regarding the main (upper) controllersof the link: the converters of DAKHLA are in DC voltage control mode, the onesat TOBENE in active power control. Both converters operate at a fixed reactivepower exchange, although the controllers can easily be modified to offer also ACvoltage control at both sides of the link.

By adding an additional control signal to the active power control loop which actson the frequency deviation in the system, the VSC converter mimics the primaryfrequency control of a classical power plant. The additional power to offer thissupport is coming from the Moroccan/European system and transferred over theDC links. This support is demonstrated by simulating an outage of KADUNA G(Nigeria, loss of 215 MW) in case the power is transferring 300 MW from TOBENEto DAKHLA. The frequency at both sides of the HVDC connection together withthe power output of the VSC at TOBENE is given in Figure 60 and Figure 61.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 124/176

Fina

lver

sion

Figure 60: Frequency at both HVDC terminals when offering frequency support after an outage of KADUNA G at t=10s

Figure 61: Active power flow over the HVDC link when offering frequency support after an outage of KADUNA G att=10sNetwork congestions within the Moroccan power system considering 1000 MW export

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 125/176

Fina

lver

sion

In the following section, it is assumed that we start from the peak load scenarioand add an additional (future) 1000 MW of solar power close to bus LAAYOUNEII.This surplus of 1000 MW is exported through the link towards the WAPP system(assume HVDC connection). Some additional assumptions have been made:

· Voltage control devices are added to the bus LAAYOUNEII (so modelled as aPV bus)

· HVDC terminal at DAHKLA is controlled to export 1000 MW and offers ACvoltage control (DAHKLA is also modelled as a PV bus)

· To simplify the modelling: HVDC is represented as a load in parallel with agenerator offering voltage control and 0 MW output (only valid for staticsimulations!)

This results in the following power flow for the case with and without the 1000 MWexport (only the part close to DAHKLA and LAAYOUNEII has been shown):

Figure 62: Power flow for an export of 0 MW – No solar pv power (purple numbers represent MW/MVar of loads, greennumbers represent MW/MVar of generation, .../... close to the busses represent bus voltage and angle, numbers close to

the lines represent the loading in %)

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 126/176

Fina

lver

sion

Figure 63: Power flow for an export of 1000 MW – 1000 MW solar pv (purple numbers represent MW/MVar of loads,green numbers represent MW/MVar of generation, .../... close to the busses represent bus voltage and angle, numbers

close to the lines represent the loading in %)

One may notice that the main influence of the export of 1000 MW is the loadingon the 400 kV double circuit from LAAYOUNEII to DAKHLA (see red circles infigure). Other parts of the network are less influenced. But even if 1000 MW istransferred, the double circuit (1070 MVA each) are only loaded by 28% or 51%(for respectively LAAYOUNEII-BOUJDOUR2 and BOUJDOUR2-DAKHLA).However, for an N-1 situation (losing for instance one circuit of BOUJDOUR2-DAKHLA, the other circuit is overloaded (116% loading).

This considered outage is repeated for different power transfers and it isconcluded that maximum 900 MW can be transferred in order to be N-1 secure.(More research is required in case the solar (pv) power is linked to bus within amore meshed part of the system. In that case, different line outages must beinvestigated for a detailed N-1 analysis.)

4.1.1.5. CONCLUSION

To link the WAPP system with the North-African system, different interconnectionoptions have been analysed and compared from a techno-economic point of view.

Taking into account the weak connection points, additional measures havetherefore to be taken to apply HVDC CSC or AC technology for a rating of 1 GVA.HVDC VSC in a bipolar configuration looks thus the most appropriate solution forthe system especially when grid support is required or future extensions (andtappings) of the system are foreseen.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 127/176

Fina

lver

sion

4.1.2. Economic studyThis study looks at the impact of interconnection with North-Africa by adding theMoroccan system to the model using the generation development plantransmitted by ONEE during the data collection and by adding a newinterconnection. This Moroccan generation park is frozen in the economicanalysis (no modification of the installed capacity). This park is shown in thefigures below at the beginning as well as at the end of the study horizon, for theyears 2017 and 2033.

At the beginning of the study period, in 2017, the generation fleet was made upmostly of coal units (43%). Then come the thermal power plants using heavy fueloil (26%), hydroelectric power plants (19%) including the pumping station-turbineof Afourer and finally the gas-fired power plants (12%).

During the period under review, investments by Morocco in renewable energiesare very important, with 5.1 GW of investment in Solar power plants Photovoltaicand 6.5 GW in wind units, these two technologies representing respectively 27%and 29% of Morocco's 2033 installeed capacity.

Figure 64: Generation installed capacity Morocco in 2017, by type of fuel

COAL; 3060;43%

NG; 834; 12%

HFO; 1903;26%

HYDRO; 1394;19%

Maroc : installed capacity 2017 (MW)

COAL NG HFO HYDRO SOLAR WIND

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 128/176

Fina

lver

sion

Figure 65: Generation installed capacity in Morocco in 2033, by type of fuel

The load load demand forecast for Morocco considered in the study is the onethat was transmitted to the consultant by ONEE during the data collection phase.The peak demand evolves from 5992 GW in 2017 to 10 882 MW in 2033, anaverage annual growth of 4%. Energy consumption goes from 37 990 GWh En2017 to 68 129 GWh in 2033, an average annual growth of 4% as well.

Based on the comparison of situations with and without interconnection, thebenefits for the West African system of interconnecting with North-Africa viaMorocco and Mauritania are estimated.

The economic study of the interconnection between North-Africa and the WAPPbegins by doing a first estimate of investment costs. Then the optimal exchangeson the interconnection are analyzed. Finally, this part of the study closes byaddressing the benefits for the West African system of interconnection with North-Africa .

It should be noted that the economic study focuses mainly on HVDC – VSCtechnology, since it was selected as the preferred solution by the technical studyand using a first analysis of the costs as is developed below.

COAL; 4005;18%

NG; 3087; 13%

HFO; 1903; 8%

HYDRO; 1045;5%

SOLAR; 6194;27%

WIND; 6510;29%

Maroc : installed capacity 2033 (MW)

COAL NG HFO HYDRO SOLAR WIND

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 129/176

Fina

lver

sion

4.1.2.1. COST ESTIMATE

A first basic cost estimation14 of the presented options can be found below (for adirect connection between DAKHLA and TOBENE). Only investment costs areconsidered in this calculation, no operational, maintenance, ... costs are included.Furthermore, the cost of the AC substations which have to be built for all studiedoptions are not included as well (for instance additional substation TOBENE400kV). Therefore, the costs as given below are merely to compare the differentoptions and give not a good estimate of the final cost.

AC (400 kV):

· Lines + compensation (400 kV double circuit): 1.5*1250 km*0.40 M$ = 750 M$· (Additional intermediate substations used for voltage control (2): 64.4 M$)· Total: Double circuit ± 814.4 M$ (Single circuit ± 564.4 M$)

HVDC (CSC - 1GW – without intermediate tapping):

· Converter stations: 2x130 M$ = 260 M$· Lines: 1250 km*0.29 M$ = 362.5 M$· Total: ± 622.5 M$

HVDC (VSC - 1GW – without intermediate tapping):

· Converter stations: 2*144 M$ = 288 M$· Lines: 1250 km*0.29 M$ = 362.5 M$· Total: ± 650.5 M$

Since the cost of an HVDC overhead line is generally less then an AC overheadline with the same capacity (a factor of 0.72 is assumed here), the cost of theconverters is compensated by the reduced line costs for long connections, seealso Figure 66. The cricitcal distiance at which there is a break-even in costbetween AC and DC is generally in the order of 600-800km. As the line is muchlonger, the above cost calculation confirms the presumption that HVDC is themost economic option for such long distance.

As indicated by the calculation, the CSC option is the cheapest option. However,when assuming that also additional investments are required to increase the SCRat the connection points, this option would possibly be as expensive or even moreexpensive as the HVDC VSC link (more detailed studies are required). Also, whencomparing the AC and HVDC options, it should be kept in mind that the AC doublecircuit is fully N-1 (1000 MVA capacity in case of an outage of one circuit), whilethe bipolar DC link can only transfer 500 MW in case of an outage (DC converteror DC line).

14 Voltage Source Converter (VSC) HVDC for Power Transmission - Economic Aspects and Comparison with other ACand DC Technologies, Cigré, Technical Brochure, 2012. & Modular Development Plan of the Pan-European Transmission System 2050: Technology assessment from 2030 to 2050, e-Highway 2050,Technical Brochure, 2014.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 130/176

Fina

lver

sion

Figure 66: Schematic illustration of the cost comparison between HVDC and AC connections (source: ABB)

4.1.2.2. ANALYSIS OF OPTIMAL ENERGY EXCHANGES

To determine the level of optimal trade between WAPP and the North-Africanpower system, the methodology consisted adding Morocco as a new node of thePRELE model developed for the analysis of the reference case presented in themaster plan above. The Moroccan generation assets were modeled according tothe information received from the ONEE during the data collection (nomodification of the installed capacity).

Then on the basis of the cost estimates made by the technical analysis, thepossibility of investing in a HVDC line between North-Africa and WAPP was alsointroduced into the economic model. The commissioning of this line is consideredfrom 2023 (the economic model could not invst before).

The purpose of optimization is to determine if there is an economic interest ininvesting in such a line, and to establish optimal flows in such case.

The results of the economic simulations indicate that there is indeed an economicinterest to interconnect the WAPP with North-Africa. These results, however, willhave to be confirmed by a detailed feasibility study that would optimize thesystems on both sides and include potential trade with Europe.

The figure here below illustrates the evolution of optimal exchanges during atypical day at the end of the study horizon, in 2033. It is observed that, most ofthe time, ernegy flows are from North-Africa to West-Africa, especially during theevening and the night during which the coal units of North-Africa are in service(Base operation), but also during the day, where the energy flows from North-Africa coming mainly of solar photovoltaic power plants. The export of solarenergy from North-Africa is explained by a better solar productible. Indeed, it ispossible to reach 2000kWh/kW in the region while the producible does notgenerally exceed 1600kWh/kW in West Africa. It should therefore be noted thatthe volume of exchange on the interconnection is limited by the availability of solarenergy in North-Africa. Thus, if aditionnal solar energy projects were installed inthat region, there would be an economic interest for West Africa to import thisaditionnal solar energy.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 131/176

Fina

lver

sion

In the evening, On the other hand, trade is the other way around, and Senegalexports the electricity produced by gas power plants, this resource being moreeconomically optimal (local ressources comparing to GNL) than the other thermaloptions available in North-Africa.

Figure 67: Optimum exchange between North-Africa and West Africa En 2033

From an energy point of view, the export from North-Africa to West-Africa areestimated at 1050 GWh per year from 2023, and to 105 GWh per year In theopposite direction, so that the net export from North-Africa to West-Africa amountsto 945 GWh annually and the optimum capacity of the link resulting from theoptimsation tool is about 260 MW.

Again, from a purely economic point of view, the flows to West Africa could besignificantly higher if solar energy was available in greater quantities in North-Africa (potentially coming from Europe). It is worth mentioning that the economicinterest of exporting wind resources available to North Africa should also bestudied as part of the feasibility study.

4.1.2.3. ANALYSIS OF IMPACTS OF INTERCONNECTION SUR THE WEST AFRICANSYSTEM

From the economic point of view, the operating cost is reduced during the day inWest Africa. The import of solar energy replaces the local thermal generationbetween 11am and 4pm. On the other hand, during the evening peak, the exportof thermal energy from Senegal increases the generation costs in the sub-region.The balance sheet is therefore relatively neutral in terms of operating costs. Note,however, that If more solar energy could be imported, Operational gains could bemore important. One could also observe a shift of solar projects from West Africato the more northerly regions, given the more abundant resources, which wouldhave the effect of reducing the capacity installed in the sub-region

-300

-200

-100

0

100

200

300

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24MW

Heure

Export (+) / Import (-) from North Africa to WestAfrica 2033

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 132/176

Fina

lver

sion

4.1.3. ConclusionThis pre-feasibility study of the interconnection between WAPP and North Africaanalysed the various possible technical possibilities for this interconnection.

On the basis of this technical analysis, it appeared that the preferred solution is aHVDC – VSC line of 1GW capacity.

The economic analysis confirmed there is an economic justification forintyerconnecting both system but with a rather limited capacity of about 260 MW.Energy flows are mainly from North-Africa to West-Africa during the night and theday, and in the opposite direction for a few hours during the evening peak.

A detailed feasibility analysis should, however, be carried out which would takeinto account the significant solar and wind potential of North Africa as well asexchange opportunities with Europe.

4.2. Interconnection opportunities with the CentralAfrican Energy Pool

After analyzing the possibilities of interconnecting WAPP with Morocco, thisscenario explores the possibility of interconnecting WAPP with the Central AfricanEnergy Pool (PEAC).

This road is part of the INGA project. The system would transfer the future powergenerated by the Inga III and Grand Inga dams and one of the export routes isWest Africa via the interconnection between Inga in DRC and Nigeria.

Concerning the Inga project, the development of the site is addressed in 7successive phases; Inga Iii Low Falls, Inga Iii High Falls, then Inga 4 to Inga 8 toreach the total generation of 42 000 MW. The first phase currently underwayconcerns Inga 3 low falls with a power of 4 800 MW.

It should be noted that, besides of Inga, other important projects, mainlyhydroelectric, should be also developed in the PEAC region and it justifies theinterests of the various interconnections studied by the PEAC.

With regard to the interconnection between the Inga site and Nigeria, the linewould be connected to Calabar, in the southeast of the country. This line is apriority project of PEAC, as well as a PIDA priority project also for 2020(infrastructure development program in Africa). The 8 countries concerned by thisinterconnection, namely the DRC, the Republic of the Congo, Gabon, EquatorialGuinea, the Cameroon, Chad, the Central Republic of Africa and Nigeria signeda memorandum of understanding. Currently, the project is in the process ofcollecting funding for studies, and few information is available concerning theexpected transfer capabilities, the volumes of energy exchanges or the tariffs tobe applied.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 133/176

Fina

lver

sion

Figure 68: Project to route the Inga-Calabar interconnection

4.2.1. MethodologyIn view of limited available information available on the charactistics of this project,the following methodology was used.

The PRELE economic model developed for the reference case has beenextended by adding, on the one hand, an area representing the PEAC and, onthe other hand, an interconnection line linking Nigeria to this new PEAC zone.

Nigeria has been able to import from PEAC Zone at a certain cost and at the levelof a certain transfer capacity considered.

· Concerning the costs of import, 3 different rates were studied: 40 USD/MWh,60 USD/MWh and 80 USD/MWh (costs including also transmission tariffs)

· Concerning the transfer capacity of the interconnection it was assumed tobe 1 GW starting from 202415 and increased to 2 GW in 2030. It should benoted that the transfer Power can only flow from the PEAC to WAPP, since thePEAC as such has not been modelled, either at the level of its load demandforecast or generation assets.

The purpose of the study is to analyze the impact on investments as well as onthe operating costs for WAPP depending on thes different rates considered.

15 This date of commissioning is taken from the regional energy Policy strategy paper Of the Central African EnergyPool

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 134/176

Fina

lver

sion

4.2.2. ResultsDifferent uses of interconnection are observed according to the tariffs applied.

§ For 40 USD/MWh, the optimization program uses the full capacity of theinterconnection is 1 GW at any time of the day until 2029. At the end of thestudy period, it uses the full capacity of 2 GW at disposal during the eveningand at night. During the day, it appears that the interconnection line is notused, given that massive investments in solar photovoltaic technology arecarried out in southern Nigeria and overall the WAPP region in the long term.

§

Figure 69: ExportAtions from the CAPP to The Wapp For a fee of 40 USD/MWh

§ For 60 USD/MWh, the line is used mainly during the evening peak between7pm and midnight and also during the night and early morning at the end ofthe study horizon (2033).

-500

0

500

1000

1500

2000

2500

1 2 3 4 5 6 7 8 9 10 1112 1314 15 1617 1819 2021 22 2324

MW

Heures

Exports du CAPP vers le WAPP40 USD/MWh

2025 2033

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 135/176

Fina

lver

sion

Figure 70: Exports from CAPP to WAPP for a fee of 40 USD/MWh (2033)

§ For 80 USD/MWh, no more imports are practically realized, except underspecial conditions, such as dry years during which Mambilla cannot deliver allof its hydroelectric potential.

In terms of impact on the investments in installed capacity, the introduction of theinterconnection of a transfer capacity of 2 GW at the end of the study is logicallyreduced the required installed capacity in peak units, mainly in Nigeria. Theinterconnection allows to ensure a reserve role and participate in the SecuritySupply system for the West African.

With regard to the impact on operational costs, the results differ according to therates considered, as summarized in the table below. With a rate of 40 USD/MWh,the intensive use of interconnection can lower Operational costs of WAPP regionof 5.5%. For 60 USD, the use of the line during the evening peak reduces thesecosts of 3.3%. For 80 USD/MWh, the impact on operational costs is virtually nil,since this line is used only in exceptional situations.

-500

0

500

1000

1500

2000

2500

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

MW

Heures

Exports du CAPP vers le WAPP60 USD/MWh

2025 2033

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 136/176

Fina

lver

sion

40 USD/MWh 60 USD/MWh 80 USD/MWh

Impact on the Investments -2 GW -2 GW -2 GW

Impact on operational costs -56 -33 -0.2%

EnergyAnnual

ExchangedGWh

1 GW Transfercapacity 5.507 2, 114 221

2 GW Transfercapacity 11.013 6.228 443

Estimated cost of imports (% ofoperational costs) 3.4% 2.3% 0.2%

Table 44: Results of the study On The interconnection of the WAPP with The PEAC

4.2.3. ConclusionAs a conclusion, it can be said that there is an economic interest in interconnectingareas of WAPP and PEAC in order to share important cheap ressources fromInga and other hydroelectric sites, notably, of the PEAC region.

The economic study carried out indicates that this interest exists provided that thepurchase price of energy is not too high. According to the first estimates, 80USD/MWh appears to be an upper limit on the tariffs to be applied. The energyexchanges and the optimal capacity are directly depending of the applied tariff.Also, beyond savings in terms of operational costs, the line also allows savings interms of investment costs, since this interconnection allows replacing someadditional thermal units, which were installed in the reference case for reasons ofreliability mainly16.

However, it is still worth mentioning that the present study considered only part ofthe problem, since the PEAC was not modelled and the fluxes were therefore onlyconsidered in the direction of the export of the PEAC to the WAPP. But it couldbe also Interesting for WAPP exporting to PEAC, which would therefore furtherstrengthen the interest of this line of interconnection.

4.3. Connection Opportunities with Cap Vert

Given its insular nature, Cape Verde is not an active member of WAPP and istherefore not studied in detail in this study. The purpose of this paragraph is topresent the country's connection options with the rest of ECOWAS.

16 Among them, the OCGT plants in Nigeria

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 137/176

Fina

lver

sion

There is no interconnected network in the country. Given the island nature of CapeVerde, the transport and distribution networks are decentralized. Thus, in eachisland transport and distribution networks are installed according to the source ofgeneration. Nevertheless, the access to electricity remains difficult in the Verdeanterritory of fragmentation of the power grid.

The project for the development of the electricity transmission and distributionsystem in 6 Islands will contribute to the improvement of technical, commercialand financial performances of The Company National of electricity (ELECTRA).The Project concerns 492 000 Inhabitants (94% of total Population of Cap Vert).It will increase The rate of Electricity access from 88% in 2010 to 98% at horizon2018.

Interconnection with the rest of ECOWAS could benefit the sub-region in twoaspects:

· On the one hand, the sharing of solar and wind resources on the island ;· On the other hand the improvement of the security of supply to Cape Verde.

Nonetheless, such a (costly) interconnection is not a priority at the horizon of thestudy. Indeed, the city of Praia, capital of Cape Verde, is located in 650km fromthe Senegalese coast. Such a distance requires using HVDC cables to connectthe two countries electrically, the cost of which is prohibitive with regard to theexchanges that could take place on this axis. Indeed, the level of demand,currently less than 100MW which is much lower than the typical transfercapabilities of the HVDC cables. On the other hand, the island's renewablepotential was estimated at 2600 MW of which 650 MW usable. The developmentof this potential should be the priority to meet the increase of national demand.

Before the interconnection with the continent, the main thing for Cape Verde istherefore(e) developing an inter-island interconnected network to improve thesecurity of supply in the territory, to better exploit the many renewable projects ofthe country and to reduce costs.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 138/176

Fina

lver

sion

APPENDIX A : GENERATION MASTER PLAN

This annex takes up the Summary of investments by country for each timehorizon, distinguishing the projects decided, the selected candidate projects. Thetime phase is finally presented.

Short term investment per country

BÉNIN

PROJECT STATUS TECHNOLOGY FUEL INSTALLEDPOWER (MW) COMMISSIONING

BID Benin Decided CCGT NaturalGas 120 2019

Parakou Decided Engine DDO 30 2019CAI Maria Gleta(Extension 50MW)

Decided CCGT NaturalGas 70 + 50 (120) 2020

PV AFD Decided PV 25 2020PV MCA SUD Decided PV 15 2020PV INNOVENTDJOUGOU Decided PV 1.5 2020

PV MCA NORTH Decided PV 30 2020PV Benin NorthStandard Selected PV 50 2021

Maria GletaWAPP Selected CCGT Natural

Gas 150 2022

Biomasse 1 Selected Biomass 10 2022Biomasse 2 Selected Biomass 11 2022Total Benin 373

Table 2: Short term invesmtents up to 2022 in Bénin

BURKINA FASO


Fada Decided Engine HFO 7.5 2018SAMENDENI Decided Hydro 2.76 2019Kossodo Decided Engine HFO 50 2020Kaya Decided PV 10 2020Zina Decided PV 26.6 2020Foot PIE 1 Decided PV 51.25 2020Zagtouli 2 Decided PV 17 2020Ouaga-Est Decided Engine HFO 100 2021Koudougou Selected PV 20 2021AFD Selected PV 50 2022PV Bobo standard Selected PV 50 2022Total Burkina Faso 385

Table 3: Short term invesmtents up to 2022 in Burkina Faso

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 139/176

Fina

lver

sion

CÔTE D’IVOIRE


Azito IV_GT Decided CCGT NaturalGas 170 2019

Ciprel V_GT Decided CCGT NaturalGas 276 2019

Korhogo Solar(RECA) Decided PV 20 2019

Azito IV_ST Decided CCGT NaturalGas 83 2020

Ciprel V_ST Decided CCGT NaturalGas 136 2021

Poro Power(CANADIANSOLAR)

Decided PV 50 2020

Centrale solaireBOUNDIALI (KFW) Decided PV 30 2020

Centrale solaireFERKE Selected PV 25 2021

BOUNDIALI 2 Selected PV 70 2021ODIENNE Selected PV 20 2021KORHOGO 2 Selected PV 30 2021SINGROBO Decided Hydro 44 2022GRIBO POPOLI Decided Hydro 112 2021Aboisso (Biokala) Decided Biomass 23 2022Aboisso (Biokala) Decided Biomass 23 2022LABOA Selected PV 100 2022FERKE 2 Selected PV 75 2022BOUTOUBRE Selected Hydro 156 2022Total Côte d'Ivoire 1443

Table 4: Short term invesmtents up to 2022 in Côte d’Ivoire

THE GAMBIA


Brikama I _G7 Decided Engine Diesel 6.4 2018Brikama III_G1 Decided Engine Diesel 10 2019Brikama III_G2 Decided Engine Diesel 10 2020World bank PV Decided PV 20 2019Brikama PV Decided PV 10 2020Standard CCGambia Selected CC HFO 60 2022

PV Gambiestandard Selected PV 50 2022

Total Gambie 166.4

Table 5: Short term invesmtents up to 2022 in The Gambia

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 140/176

Fina

lver

sion

GHANA


KPONT_ST Decided CCGT NaturalGas 120 2018

CENPOWER_CC Decided CCGT NaturalGas 360 2018

EARLY POWER Decided CCGT NaturalGas 147 2019

TROJAN 3 Decided OCGT NaturalGas 50 2018

MI ENERGY Decided PV 20 2018BUI PHASE 1 Decided PV 50 2018

GPGC Decided CCGT NaturalGas 170 2019

EARLY POWER Decided CCGT NaturalGas 153 2019

KALIOU LORA Decided PV 12 2019BIO THERM Decided PV 20 2019

AMANDI Decided CCGT NaturalGas 240 2020

BUI PHASE 2 Selected PV 200 2021BONGO SOLAR Selected PV 40 2021

ROTAN Decided CCGT NaturalGas 330 2022

PV Ghana northStandard Selected PV 250 2022

Total Ghana 2162

Table 6: Short term invesmtents up to 2022 in Ghana

GUINÉA


ENDEAVOR Decided Engine DDO 50 2019Khoummaguély PV Decided PV 40 2019Sougéta PV Decided PV 30 2019KALETA extensionreservoir Decided Hydro 0 2021

SOUAPITI Decided Hydro 450 2020PV Guinea Northstandard Selected PV 100 2021

PV Guinea SouthEast standard Selected PV 100 2021

FOMI Decided Hydro 90 2022KOGBEDOU Decided Hydro 58 2022FRANKONEDOU Decided Hydro 22 2022Touba Decided Hydro 5 2022Touba PV Decided PV 5 2022Total Guinea 950

Table 7: Short term invesmtents up to 2022 in Guinea

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 141/176

Fina

lver

sion

GUINÉA-BISSAU


Bor (BOAD) Decided Engine HFO 15 2019BADEA Decided Engine DDO 22 2019Standard CCGuinee Bissau Selected CC HFO 60 2022

PV Guinea BissauStandard Selected PV 50 2020

Total GuineaBissau 147

Table 8: Short term invesmtents up to 2022 in Guinea Bissau

LIBÉRIA


Extensionreservoir MountCoffee

Selected Hydro 0 2022

PV Liberiastandard Selected PV 50 2022

Total Liberia 50

Table 9: Short term invesmtents up to 2022 in Liberia

MALI


ALBATROS(Kayes) Decided Engine HFO 92 2018

Bamako (Sirakoro) Decided Engine HFO 100 2020GOUINA Decided Hydro 140 2020KITA PV Decided PV 50 2020Kati PV Decided PV 65 2021Segou PV Decided PV 33 2021Sikasso PV Selected PV 50 2022Total Mali 530

Table 10: Short term invesmtents up to 2022 in Mali

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 142/176

Fina

lver

sion

NIGER


Malbaza PV Decided PV 7 2019Lossa PV Decided PV 10 2019NCE (Maradi) PV Decided PV 30 2019Zinder PV Decided PV 60 2019Dosso PV Decided PV 10 2019Niamey PV Decided PV 30 2019

Goroubanda PV 2 Decided PV 30 2019

GOROUBANDA 2 Decided Engine HFO 20 2020Diesel North Decided Engine HFO 6 2020

Goroubanda PV 1 Decided PV 20 2020

Agadez PV Decided PV 13 2020SALKADAMNAphase 1_1 Decided Charbon 50 2021

SALKADAMNAphase 1_2 Decided Charbon 50 2021



KANDADJI Decided Hydro 130 2021PV standardNiamey Selected PV 100 2022

PV standard NigerNorth Selected PV 100 2022

Total Niger 766

Table 11: Short term invesmtents up to 2022 in Niger

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 143/176

Fina

lver

sion

NIGERIA


EGBEMA I - NIPP Decided OCGT NaturalGas 113 2018

OMOKU - NIPP Decided OCGT NaturalGas 113 2018

AZURA Decided OCGT NaturalGas 450 2018

AFAM III Decided OCGT NaturalGas 240 2018

EGBEMA I - NIPP1 Decided OCGT Natural

Gas 113 2019

EGBEMA I - NIPP2 Decided OCGT Natural

Gas 113 2019

KADUNA IPP Decided OCGT NaturalGas 215 2019

OMOKU - NIPP Decided OCGT NaturalGas 113 2019

KASHIMBILLA Decided Hydro 40 2019OKPAI IPP II -AGIP 1 Decided CCGT Natural

Gas 300 2020

OKPAI IPP II -AGIP 2 Decided CCGT Natural

Gas 150 2020

Nova solar Selected PV 100 2021

Nova scotia power Selected PV 80 2021

Pan africa solar Selected PV 75 2021

Lr aaron solarpower plant Selected PV 100 2021

Egbin 2+ phase 1 Selected CCGT NaturalGas 1200 2022

ZUNGERU Selected Hydro 700 2022Quaint energysolutions Selected PV 50 2022

Nigeria solarcapital partners Selected PV 100 2022

Afrinergia solar Selected PV 50 2022PV Nigeria EastStandard Selected PV 150 2022

Total Nigeria 4565

Table 12: Short term invesmtents up to 2022 in Nigeria

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 144/176

Fina

lver

sion

SÉNÉGAL


Diass Decided PV 25 2018Wind turbine 1 Decided Wind Turbine 50 2019Touba (scalingsolar) Decided PV 30 2019

Kaloack (scalingsolar) Decided PV 30 2019

Sendou IPP CES I Decided ST Charbon 115.1 2020

Malicounda Decided Engine HFO 120 2020

SAMBANGALOU Decided Hydro 128 2022

Wind turbine 2 Decided Wind Turbine 50 2020Wind turbine 3 Decided Wind Turbine 50 2021Scaling solar Selected PV 40 2021

Kayar Kounoune Decided CCGT NaturalGas 115 2022

World Bank project Selected PV 100 2022PV Dakar Standard Selected PV 50 2022PV TambacoundaStandard Selected PV 50 2022

Total Sénégal 953

Table 13: Short term invesmtents up to 2022 in Sénégal

SIERRA LEONE


Newton Solar Decided PV 6 2019CEC Africa Phase1 Decided Engine HFO 50 2020

Heron Energy Selected PV 5 2021Bo PV Selected PV 5 2021PV FreetownStandard Selected 100 2022

Total Sierra Leone 166

Table 14: Short term invesmtents up to 2022 in Sierra Leone

TOGO


Lome TG Decided OCGT Natural Gas 60 2020PV Dapaong Selected PV 30 2021Total Togo 90

Table 15: Short term invesmtents up to 2022 in Togo

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 145/176

Fina

lver

sion

Medium term investments per country

BÉNIN


Maria Gleta WAPP Selected GT (CC) NaturalGas 150 2023

Maria Gleta WAPP Selected ST (CC) NaturalGas 150 2024

GREENHEART POWERAFRICA Selected PV 10 2025

PV Benin North Standard Selected PV 150 2024-2026Total Benin 460

Table 16: Medium term investments 2023-2029 in Bénin

BURKINA FASO


WAPP PV Ouagadougou Selected PV 150 2022-2024

PV Bobo standard Selected PV 50 2025Foot PIE 2 Ouagadougou Selected PV 100 2026PV Ouaga standard Selected PV 150 2029Projet Wind Ouaga Selected Wind Turbine 75 2029Total Burkina 525

Table 17: Medium term investments 2023-2029 in Burkina Faso

CÔTE D’IVOIRE


WAPP PV CIV Selected PV 150 2022-2024PV CIV Northstandard Selected PV 150 2022-2024

Louga Decided Hydro 224 2023TIBOTO Decided Hydro 112.5 2028San Pedro I_ST1 Decided ST Charbon 350 2026

PV CIV Northstandard Selected PV 100 2027

PV CIV Sudstandard Selected PV 100 2028

San Pedro I_ST2 Decided ST Charbon 350 2029

Total Côte d'Ivoire 1536

Table 18: Medium term investments 2023-2029 in Côte d’Ivoire

NORTH


WAPP PV Gambie Selected PV 150 2023-2025Total Gambie 150

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 146/176

Fina

lver

sion

Table 19: Medium term investments 2023-2029 in The Gambia

GHANA


PV Ghana northstandard Selected PV 250 2023

PV Ghana northstandard Selected PV 250 2025

WAPP PV Ghana Selected PV 150 2026PV Ghana northstandard Selected PV 250 2028

PV Ghana sudstandard Selected PV 100 2029

CCGT Aboadze Selected CCGT NaturalGas 450 2029

Total Ghana 1450

Table 20: Medium term investments 2023-2029 in Ghana

GUINÉA


AMARIA Decided Hydro 300 2023MORISANAKO Selected Hydro 100 2025GRAND KINKON Selected Hydro 291 2023KOUKOUTAMBA Decided Hydro 294 2024BONKON DIARIA Selected Hydro 174 2025TIOPO Selected Hydro 120 2028PV Guinea SouthEast Selected PV 100 2028

DIARAGUÔLA Selected Hydro 72 2029Total Guinea 1451

Table 21: Medium term investments 2023-2029 in Guinea

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 147/176

Fina

lver

sion

GUINÉA-BISSAU


PV Guinea Bissau Selected PV 50 2024PV Guinea Bissau Selected PV 50 2028Total GuineaBissau 100

Table 22: Medium term investments 2023-2029 in Guinea Bissau

LIBERIA


PV Liberiastandard Selected PV 50 2025

Total Liberia 50

Table 23: Medium term investments 2023-2029 in Liberia

MALI


Kurikolo PV Selected PV 50 2023Koutiala PV Selected PV 25 2024WAPP Regional Project Selected PV 150 2022-2024Fana PV Selected PV 50 2025PV Bia Selected PV 40 2026Tenkele PV Selected PV 40 2027Medium PV Selected PV 40 2028Total Mali 395

Table 24: Medium term investments 2023-2029 in Mali

NIGER


PV standard Niamey Selected PV 100 2024PV standard North Selected PV 150 2025Projet wind standardNiamey Selected Wind Turbine 150 2026

PV standard Niamey Selected PV 100 2027PV standard North Selected PV 100 2028PV standard Niamey Selected PV 100 2029Total Niger 700

Table 25: Medium term investments 2023-2029 in Niger

NIGERIA


WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 148/176

Fina

lver

sion

EGBIN 2+ 2 Selected ST (CC) Natural Gas 700 2023

ETHIOPE 1 Selected GT (CC) Natural Gas 344 2023

MABON Selected Hydro 39 2023

KVK power nigeria LTD PV Selected PV 55 2023Anheed kafachan solar IPPPV Selected PV 100 2023

CT cosmos PV Selected PV 70 2023

Oriental PV Selected PV 50 2023

ETHIOPE 2 Selected ST (CC) Natural Gas 156 2024

CALEB INLAND Selected CCGT Natural Gas 500 2024

ETHIOPE 3 Selected GT (CC) Natural Gas 344 2024

MAMBILLA Decided Hydro 3050 2024

EN consulting - Kaduna Selected PV 100 2024Kazure (Kano disco) phase1 Selected PV 500 2024

ETHIOPE 4 Selected ST (CC) Natural Gas 156 2025

CALEB INLAND 2 Selected CCGT Natural Gas 500 2025

ALAOJI 2+ NIPP Decided CCGT Natural Gas 285 2025Standard CC Nigeria South1 Selected CCGT Natural Gas 450 2026

Standard CC Nigeria South2 Selected CCGT Natural Gas 450 2026

Standard CC Nigeria South3 Selected CCGT Natural Gas 450 2026

Motir dusable Selected PV 100 2026

Middle band solar Selected PV 100 2026

WAPP PV Nigeria Selected PV 1000 2025-2029

GEREGU NIPP 2 Selected ST (CC) Natural Gas 285 2027

OMOTOSHO II 2+ Selected ST (CC) Natural Gas 254 2027

CALEB INLAND 3 Selected CC Natural Gas 500 2027Standard CC Nigeria South4 Selected CCGT Natural Gas 450 2028

Standard CC Nigeria South5 Selected CCGT Natural Gas 450 2028Standard CC Nigeria South6 Selected CCGT Natural Gas 450 2028

Projet Wind standard North Selected Wind Turbine 350 2028

GEREGU FGN1-2 Selected GT (CC) Natural Gas 414 2029CALABAR / ODUKPANI -NIPP Selected ST (CC) Natural Gas 254 2029

GBARAIN / UBIE 2 Selected ST (CC) Natural Gas 115 2029Standard CC Nigeria South7 Selected CCGT Natural Gas 450 2029

Total Nigeria 13471

Table 26: Medium term investments 2023-2029 in Nigeria

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 149/176

Fina

lver

sion

SÉNÉGAL


Standard PVDakar Selected PV 100 2023


Standard CCGTSénégal Selected CCGT Natural

Gas 450 2025

Standard CCGTSénégal Selected CCGT Natural

Gas 300 2025


Total Sénégal 1050

Table 27: Medium term investments 2023-2029 in Sénégal

SIERRA LONE


BUMBUNA II Decided Decided Hydro 132 2023BUMBUNA III (Yiben)Decided Decided Hydro 66 2023

BENKONGOR ICandidate Selected Hydro 34.8 2023

BENKONGOR IICandidate Selected Hydro 80 2025

BENKONGOR IIICandidate Selected Hydro 85.5 2026

PV Standard SierraLeone Selected PV 50 2023

PV Standard SierraLeone Selected PV 50 2028

Total Sierra Leone 498

Table 28: Medium term investments 2023-2029 in Sierra Leone

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 150/176

Fina

lver

sion

TOGO


Sarakawa Decided Hydro 24.2 2023PV Blitta Selected PV 20 2023Adjarala Decided Hydro 147 2026WAPP PV Togo Selected PV 150 2028-2030Total Togo 341

Table 29: Medium term investments 2023-2029 in Togo

Long term investments per country

BÉNIN


Standard CC BeninNorth Selected OCGT HFO 60 2030

PV Benin SudStandard Selected PV 100 2030

Total Benin 260

Table 30: Long term investments 2030-2033 in Benin

BURKINA FASO


Standard CC Bobo-Dioulasso Selected OCGT HFO 60 2029

PV Ouaga standard Selected PV 150 2030Total Burkina 210

Table 31: Long term investments 2030-2033 in Burkina Faso

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 151/176

Fina

lver

sion

CÔTE D’IVOIRE


CC Songon Selected CC NaturalGas 369 2031

Standard TAG Côted'Ivoire Sud Selected OCGT Natural

Gas 300 2030

PV CIV Sudstandard Selected PV 250 2030

Standard CC Côted'Ivoire North Selected OCGT HFO 60 2033

Total Côte d'Ivoire 979

Table 32: Long term investments 2030-2033 in Côte d’Ivoire

GHANA


PV Ghana Sudstandard Selected PV 300 2030-2033

Projet EolienStandard GhanaNorth

Selected Wind Turbine 200 2030

Standard TAGGhana Sud Selected OCGT GN 300 2033

Total Ghana 800

Table 33: Long term investments 2030-2033 in Ghana

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 152/176

Fina

lver

sion

GUINEA


BOUREYACandidate Selected Hydro 160 2030

PV Guinea Northstandard Selected PV 100 2030Standard CC GuineaNorth Selected OCGT HFO 60 2030

FETORE Selected Hydro 124 2031LAFOU Selected Hydro 98 2032Total Guinea 542

Table 34: Long term investments 2030-2033 in Guinea

GUINEA-BISSAU


Standard CC GuineaBissau Selected OCGT HFO 60 2030

Total Guinea Bissau 60

Table 35: Long term investments 2030-2033 in Guinea Bissau

LIBERIA


PV Libéria standard Selected PV 50 2030Standard CC Libéria Selected OCGT HFO 60 2030Mano Selected Hydro 180 2032Total Libéria 290

Table 36: Long term investments 2030-2033 in Liberia

MALI


Standard CC Mali Selected OCGT HFO 60 2030PV Mali Bamakostandard Selected PV 100 2031

Total Mali 160

Table 37: Long term investments 2030-2033 in Mali

NIGER


PV Niger Niameystandard Selected PV 150 2030

PV Niger Northstandard Selected PV 150 2030

Projet EolienStandard NigerNiamey


Standard CC Niger Selected OCGT HFO 60 2030

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 153/176

Fina

lver

sion

Projet EolienStandard NigerNorth


Total Niger 560

Table 38: Long term investments 2030-2033 in Niger

NIGERIA


EGBEMA II Selected ST (CC) Natural Gas 127 2030IHOVBOR (EYAEN) 2- NIPP Selected ST (CC) Natural Gas 254 2030

Standard CC Nigeria Selected CC Natural Gas 1500 2030Standard OCGTNigeria Selected OCGT Natural Gas 500 2030Kazure (Kano disco)phase 2 Selected PV 500 2030PV Nigeria Southstandard Selected PV 400 2030-2033

Standard CC Nigeria Selected CC Natural Gas 2500 2031Standard OCGTNigeria Selected OCGT Natural Gas 1000 2031



Total Nigeria 13781 2033

Table 39: Long term investments 2030-2033 in Nigeria

SÉNÉGAL


Standard CCSénégal Selected CC Natural

Gas 450 2030

Standard CCSénégal Selected CC Natural

Gas 450 2030

PV Dakar standard Selected PV 150 2031PV Wind Dakarstandard Selected Wind Turbine 150 2031PV Wind Dakarstandard Selected Wind Turbine 200 2033

Total Sénégal 1400

Table 40: Long term investments 2030-2033 in Sénégal

TOGO


Standard CC TogoNorth Selected OCGT HFO 60 2030

Total Togo 60

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 154/176

Fina

lver

sion

Table 41: Long term investments 2030-2033 in Togo

ANNEXE B: TRANSMISSION MASTER PLAN

National Reinforcements – 2022

Senegal

· Line 225 KV Mbour Fatick Kaolack and substationFatick 225 KV· New Line 225 KV double circuit Sendou Kounoune· New Line 225 KV double circuit Tobene Kounoune· New Cable Underground 225 KV double circuit Patte D’Oie Kounoune and

225 kV substation Patte D’Oie· New Line 225kV Tambacounda Kolda Ziguichor· New Line 225kV Damniado - Aprosi and creation of the two substations in

225kV· New Line 225kV Double circuit Tobène Saint-Louis Nouakchott and related

substations· New 225 kV line Tobene - Thies – Diass (decided project)· OMVG Line and associated substations

The Gambia

· New 132 kV double circuit line Brikama - Jabang - Kotu· Two new 225/132 kV transformers in Brikama· OMVG line and associated substations

Guinea Bissau

· OMVG line and associated substations

Guinea

· Second circuit 225 kV Kaleta - Linsan· Second SVC 15 MVAr at Linsan· Two new transformers 225/110 kV at Linsan· OMVG line and associated substations· CLSG line and associated substations

Mali

· Bamako loop : New double circuit line 225 kV Sikasso - Bougouni -Sanakoroba - Dialakorobougou - Kenie - Banconi - Kati - Kodialani –Sanakoroba

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 155/176

Fina

lver

sion

· Mining loop – New double circuit line 225 kV Kayes - Diamou - Sadiola -Loulo - Manantali

· New double circuit line 225kV Manantali – Bamako through Kita and Kati· New substation 150kV at Dialakorobougou on break in of line Fana-Segou· New transformer 225/150 kV Kodialani

Sierra Leone

· CLSG line and associated substations· New 225 kV line Yiben - Waterloo and Waterloo 225kV substation· New 161 kV line Waterloo - Freetown and 2 new 225/161 kV transformers

in Waterloo

Liberia

· CLSG line and associated substations

Côte D’Ivoire

· New Line 400kV Bakre - Akoupe Zeudji PK24· New Line 400 KV Azito IV- Bakre· New Line 400 KV Azito IV- Akoupe Zeudji PK24· New Line 400 KV Bakre -Bingerville· New Line 400 KV Akoupe Zeudji PK24 -Bingerville· Two new Transformers 400/225 KV Akoupe Zeudji PK24· Three new Transformers 400/225 KV Bingerville· Second line 225 KV San Pedro- Soubre passing through Boutoubre and

Gribo Popoli· New 225kV substation Yopougon 1 in cut-in of Abobo-Azito· Upgrade of lines 90 kV Treichville-Vridi to 225 kV· New Line 225 KV Akoupe Zeudji PK24 -Yopougon 3-Azito· New Line 225 KV Boundiali- Tengrela· Second circuit 225 KV Vridi- Bia Sud - Riviera· New substation 225 KV Bia South in cut-in of Vridi-Riviera line· New Line 225 KV Akoupe Zeudji PK24 Anyama Adzope - Attakro Daoukro

- Serebou -Dabakala-Kong- Ferke· New Line 225 KV Bouake- Serebou -Bondoukou· New Line 225 KV double circuit Bingerville- Anani and substation Anani

225 KV· New substation 225 KV Gagnoa and Divo· Break in the second line Taabo -Abobo by Akoupe Zeudji PK24· New Line 225 KV Laboa -Boundiali-Korhogo- Ferke· New Line 225 KV Buyo - Dueko - Man· New Line 225 KV Dueko - Zagne - Toulepleu· Second circuit Buyo - Soubre 225 KV· New Line 225kV Taabo -Yamoussoukro-Kossou· New Line Bouake-Bouake 3-Kossou and new substation Bouake 3· New Line 225 KV Buyo -Daloa· New substation Katiola 225 KV in line break Ferke – Bouake

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 156/176

Fina

lver

sion

· CLSG Line and associated substations

Ghana

· New 330kV line Bolgatanga - Tamale - Kintampo - Kumasi - Dunkwa –Aboaze and associated 330 kV substations and 330/161 kV transformers

· New 330 kV substation Dawa and B5+ on the line Volta - Davié· Two new 330/161 kV transformers at Dunkwa· New 161 kV line Yendi - Juale - Kadjebi - Kpandu – Asiekpe (Convertion

of the existing 69 kV line Kadjebi - Kpandu to 161 kV)· Upgrade of 161 kV line Takoradi - Tarkwa to 364 MVA· New Accra Central substation and update 2 circuits Volta-Accra East-

Achimota-Accra Central-Mallam to 488 MVA· New 161 kV substation at Berekurum· New 330 kV line Karpower - Aboaze· New 161 kV double circuit line Aksa - Smelteri II

Niger

· Two new 330/132 kV transformers Goroubanda· Nouvelle ligne 330kV double terne Gorou Banda - Salkdamna· Nouvelle ligne 132 kV double terne Kandadji - Niamey· Nouvelle ligne 132 kV Salkadmna - Tahoua - Keita - Malbaza· Remplacement de la ligne 132 kV Gazaou - Zinder vers 109 MVA

Burkina

· New Line 132 KV Zano - Koupela and substation Koupela 132 KV· New SVC 50 MVAr decided at Pa· 3 Transformers 330/225 KV at Ouaga East· New double circuit line 225 kv Ouaga East-Ouaga South East and 225/132

KV Ouaga South East substation in break of line 132 kv Patte D’Oie - Zano· New Line 225 KV Ouaga South East-Ouaga South· New Line 225 KV Ouaga south- Zagtouli In Parallel Of the Ghana-Burkina

Faso interconnection, which breaks in at Ouaga Sud

Benin

· New Line 161 KV Malanville - Bembereke and related substations (Malanville- Guene - Kandi – Bembereke)

· Break in of 20 Km of the Maria Gleta CCGT on the 330 KV line Ghana-Nigeria· New double circuit line 161 KV Onigbolo – Parakou· New Line 330 KV Davié – Sakete· New Line 161 KV Benin-Togo and related substations (Dapaong Mandou -

Porga - Tanguieta – Natitingou)· Upgrade of 161 KV conductors Mome Hagou - Maria Gleta To 178 MVA

Togo

· New Line 330 KV Davié - Sakete· New Line 161 KV Ghana North-Togo and associated substations (Bawku -

Cinkasse -Dapaong-Mango – Kara)

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 157/176

Fina

lver

sion

· New Line 161 KV Benin-Togo and related substations (Dapaong Mandou -Porga - Tanguieta – Natitingou)

· New 161 KV substation Notse and 161kV Line Notse-Atakpame and line 161KV Notse - Davié

· New substation 161 KV Legbassito and 161kV line double circuit Legbassito- Davié

· Replacement of 161 KV conductors from Mome Hagou - Maria Gleta to 178MVA

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 158/176

Fina

lver

sion

Nigeria

The recent master plan from TCN was utilized as the main source forreinforcements and it was agreed that the 2022 model would be based on thesituation described in the TCN master plan for the study year 2020.

· Second circuit line Benin - Omotosho - Ikeja West – Egbin· New 330 kV substation at Katsina and two 330/132 kV transformers· New double circuit 330 kV line Katsina – Kano· New double circuit 330 kV line Akangba – Alagbon· New double circuit 330 kV line Onitsha - Nnewi - Owerri - Egbema - Omoku· New double circuit 330 kV line Alaoji - Owerri· New single circuit 330 kV line Osogbo - Akure - Ihovbor· Two new 330/132 kV transformers in Akure· New Epe substation and new double circuit 330 kV line Aja - Epe - Omotosho

Second circuit line Benin - Omotosho - Ikeja West - Egbin

· Second circuit line Egbin - Aja· New 330 kV substation at Port Harcourt and two 330/132 kV transformers· New double circuit 330 kV line Delta - Port Harcourt - Afam - Ikot Ekpene - Ikot

Abasi· New 330 kV substation Ikot Abasi and three 330/132 kV transformers· New double circuit 330 kV line Lokaja - Obajana· New double circuit 330 kV Gwagwalada - Eastmain and associated substation

and transformer· New double circuit 330 kV line Katsina – Kazaure – Dutse - Bauchi· Second circuit 330 kV line Katampe - Shiroro· Second 330 kV line Kaduna - Kano· New 330 kV substation at Zaria and connection with a single circuit 330 kV line

to Kano and to Kaduna and two new 330/132 kV transformers in Zaria· Second and third 330 kV line Kaduna - Jos and two new 330/132 kV

transformer at Jos· New 330 kV substation at Katsina and two 330/132 kV transformers· New double circuit 330 kV line Katsina - Kano· Two new 330/132 kV transformers in Damaturu· New 330/132 kV transformer in Asaba· New 330/132 kV transformer in Benin· Extension of the 132kV lines such as proposed in the TCN Master Plan· Reconductoring of 28 132 kV lines to a higher capacity· New 330 kV substation at Bauchi· New single circuit 330 kV line Jos - Bauchi - Gombe· New 330 kV substation at Abakaliki and two new 330/132 kV transformers· New double circuit 330 kV line Ugwaji - Abakaliki· New double circuit 330 kV line Akangba - Alagbon· New 330/132 kV transformers in Delta, Osogbo, Katampe, Ajaokuta(x2),

Omoku, Ayede and New Haven· New 330 kV substation Lafia on the line Jos-Makurdi and two new 330/132 kV

transformers

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 159/176

Fina

lver

sion

· New 330 kV substation Aliade on Ugwaji-Makurdi and new 330/132transformer


Senegal

· Second OMVG line Kaolack to Brikama

The Gambia

· Second OMVG line through Brikama

Guinea Bissau

· Second OMVG line through Bissau - Mansoa - Bambadinca - Salthinho

Guinea

· New double circuit 225 kV Maneah - Linsan· New circuit 225 kV Amaria - Kaleta· New loop and associated substations: Faranah - Kissidougou - Guekedou

- Macenta - Nzerekore· New 225 kV interconnection circuit Fomi - Morisanako - Boundiali· New 225 kV double interconnection circuit Linsan - Koukoutamba -

Boureya - Manantali· New 225 kV double circuit Labé-Koukoutamba· New 110 kV substation of Sonfon in between Matoto - Tombo

Mali

· New 225 kV circuit Sikasso - Syama· New 225 kV circuit Koutiala - San - Mopti· Increase of the voltage level of the circuit Segou - Fana – Dialakorobougou· from 150 kV to 225 kV· New 225/150 kV transformers at Fana et Dialakorobougou· New 225/150 kV transformer Kodialani· Second 150 kV circuit Sirakoro - Dialakorobougou

Sierra Leone

· Second circuit of 225 kV line Yiben - Waterloo· New Porto Loko 225/161 substation on Yiben Waterloo line and new 161kV

line to Lunsar· New 225 kV line Waterloo - Moyamba - Lanti - Bo - Baomahun· Second 225/161 kV transformer Bumbuna and upgrade of the first

transformer

Liberia

· New 225/66 kV transformers at Monrovia

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 160/176

Fina

lver

sion

Côte D’Ivoire

· New substation at Grand Bassam 225 kV and· adjecent 225 kV circuit Anani-Grand Bassam· Second circuit on the 225 kV line Anani-Grand Bassam· Second circuit on the 225 kV line Bondoukou - Serebou

Ghana

· New 161 kV line New Aberim - Akwatia· Second circuit 161 kV Volta - Kpong· Change conductor of 161 kV line CapeCoast - Aboaze to 488 MVA· Change conductor of 161 kV kine Dunkwa-New Obuasi to 364 MVA· New 330/161 kV transformer in Volta· Change conductor of 161 kV kine Akosombo - Asiekpe to 364 MVA· New 161 kV substation Atebubu and associated lines· New 161 kV substation Salaga and Kete-Krachi and associated lines

Burkina Faso

· Upgrade of 225 kV interconnection Bolgatanga – Ouaga to double circuit· New 225/90 kV transformers at Pa

Benin

Togo

· New 161 kV line Kara - Badjeli· New 161 kV line Kara - Atakpame

Niger

· New 330 kV double circuit Salkadamna - Sonichar and 2 new 330/132 kVtransformers at 330kV substation of Sonichar

Nigeria

· New double circuit 330 kV line Wukari Lafia - Apo· New double circuit 330 kV line Obajana - Ganmo· New 330/132 kV transformers at EastMain, Alagbon, Omotosho (x2),

Onitsha, Lokaja, Ihiala, Gombe


Senegal

· New 225 kV line Tambacounda-Bakel· New 225 kV line Matam 2- Linguere -Touba· New Transformer 225/90 KV Tobene· New 225 kV substation at Cap des Biches on break in of line Kounoune –

Patte D’Oie· Second circuit of 225 kV line Tobene Sakal

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 161/176

Fina

lver

sion

· New Transformers 225/90 kV in Kounoune· New 225 kV double circuit line Mboro-Tobene

Guinea

· New 225 kV line double circuit Maneah - Matoto and three transformers225/110 KV

· New 225 kV substation in Boureya

Mali

· New 225 kV interconnection line Tengrela - Syama

Côte d’Ivoire

· New 400 kV substation in San Pedro and 2 transformers 400/225 KV New400 kV line double circuit San Pedro- Akoupe Zeudji (PK24)

· New 400 kv line San Pedro to Man and two transformers 400/225 KV inMan

· New double circuit line 225 kV Yopougon 3-Songon· New 225 kV line San Pedro- Tiboto - Buchanan· New transformer 400/225 kV at Akoupe Zeudji· Second circuit 225 KV Yopougon 3-Azito· New 225 kV line Daloa – Kossou· New 225 kV interconnection line Tengrela - Syama

Ghana

· New Pokuase 330/161 KV substation and associated transformers· New 161 kV line Pokuase - Mallam· Second Circuit of 330 KV line Aboaze - Dunkwa -Kumasi and new 330/161

KV transformers· New 330 KV line Kumasi- Pokuase· New 330 KV line Bolgatanga- Juale - Dawa and new 330/161 KV

transformers at Bolgatanga, Dawa and Juale· Change Conductor of 161 KV line Dunkwa-Ayanfuri-Asawinso to 364 MVA· New 330/161 KV transformers in Bolgatanga, Kumasi and Kintampo (x2)· New 161 KV Substation Atebubu and associated lines· New 161 KV substations Salaga and Kete-Krachi and associated lines

Burkina Faso

· New 225 kV substation Ziniare and transformers 225/90 kV associated· New line 225 KV Ouaga East Ziniare - Zagtouli· Second circuit of the line Ouaga South East-Patte D’Oie 132 kV· New Transformers 225/132 KV Ouaga south East· New double circuit interconnection line Bolgatanga-Bobo· Second line 225 KV Pa-Bobo

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 162/176

Fina

lver

sion

Benin

· New Transformer 330/161 kV Sakete· New Line 161 KV Adjarala -Bohicon· New double circuit line 161 KV Adjarala - Avakpa

Togo

· New Transformer 330/161 KV Davié· New Line 161 KV Adjarala - Nangbeto· New Line 161 KV Adjarala -Bohicon· New Line 161 KV Adjarala - Notse· New Line 161 KV Adjarala - Mome Hagou· New double circuit line 161 KV Adjarala - Avakpa

Niger

· New Line 330 KV Salkadamna - Goudel Gorou and transformer 330/132KV at Goudel Gorou

· New Line 330 KV double circuit Salkadamna - Malbaza - Gazoua -Katsinaand related substations

· Two new transformers 330/132kV Goroubanda

Nigeria

· New 330/132 KV Transformers in Ikeja, Jos, Ganmo, Egbin, Benin Ayede,Delta, Katampe, Adiabor, New Agbara, Akangba

· New 330 KV Line Owerri Egbema· New 330 KV Line Geregu – Ajaokuta· New 330 KV Line Ikeja- Akangba – Omotosho· New 330 KV Line Gwagwadala – Katampe· New 330/132 KV transformers in Delta and Abakaliki· New Dual circuit 330 KV line New Agbara - Akangba· New 330 KV Line Yola-Gombe

Development of the dynamic model

GENERATING UNITS

The dynamic models of generating units have been created from the informationcollected from each Country and from the data already available to the Consultant.Each model includes the following parts:

· Parameters of the alternator: GENSAL model for salient pole machine andGENROU model for round rotor units;

· Type and parameters of the Governors: several types of governor areimplemented in the model, depending on the characteristics of each unit.

· Type and parameters of the Exciters: several types implemented dependingon the structure of the excitation systems and its controllers, according to IEEEstandardization.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 163/176

Fina

lver

sion

Particular attention has been put on combined cycle (CC) units. The selected CCmodel represents each component unit as a single machine, matched with eachother through output power instead of exhaust flows. This simplification isacceptable given the uncertainties present in planning studies.

The solar photovoltaic plants have been modelled through an aggregatedstandard model based (PVGEN80).

The wind farms have also been represented by an aggregated model for eachwind farm (WINDFEQ).

DYNAMIC LOAD MODEL

For producing realistic results, two load models have been used in dynamicsimulations, taking into account the behaviour of the distribution network. Thismodel represents loads aggregated at the MV voltage level.

A key factor for load modelling is the proportion of induction motors, whichrepresent rotating loads. It is assumed that induction motors amount to 40% ofthe load.

Therefore, two types of load model will be considered: an impedance model (60%)and a "distribution network" type model (40%).

· Impedance type load model

The impedance load model is frequently used to represent the response ofactive and reactive power to frequency and voltage variations. Mathematically,the behaviour of this load model is described by the following equations:

( ) = .( )

.( )

( ) = .( )

.( )

Where a, b, c and d are constants whose values are set depending on the typeof load (residential, industrial…). In the framework of this study, the followingassumptions will be taken:

- Active/reactive power varies with the square of the voltage (a=b=2)- Variation of active/reactive power with frequency are neglected (c=d=0)

· "Distribution network" type load model

This model represents the distribution network downstream of the HV/MV step-down transformer, characterised by a significant proportion of rotating loads.

Figure 71 shows the structure of the distribution network type model.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 164/176

Fina

lver

sion

Figure 71: Distribution network type load model

The model includes:

· A step-down transformer with a continuous under load tap changer.· A distribution cable modelled by an impedance.· A shunt compensator connected to the secondary of the transformers.· A generic induction motor connected at the end of the distribution cable.· A resistive load connected at the end of the distribution cable.

The shunt compensator is adjusted to align the active power absorbed by the loadmodel with the results of the static simulations. Standard values of small inductionmotors are used for the parameters of the model.

Transformers

Min turn ratio [pu] 0.9

Max turn ratio [pu] 1.22

Time constant [s] 20

TFO loading [%] 60

Nominal voltage [pu] 1.03

Resistance [pu] 0.005

Leakage Reactance [pu] 0.035

Feeder

Voltage drop [pu] 0.01

X/R [-] 0.5

Load Mix

Loading of motors [%] 100

Rotating load [%] 100

Inertia H [MW.s/MVA] 0.5

efficiency [-] 95

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 165/176

Fina

lver

sion

Nominal mech power [pu] 0.87

Starting torque [pu] 0.77

Maximum torque [pu] 2.3

Nominal speed [t/min] 2959

Starting current [pu] 5.6

Table 45: Step-down transformers parameters in distribution network type load model

EXISTING AND PLANNED PSS SCHEME

The dynamic model includes generic PSS models, damping a wide range offrequencies, installed in various power plants in the WAPP system according tothe information collected from the Client, as presented in the Table 46.

Country Power Plant Type # Units

Cote d'Ivoire Ciprel Thermal 7

Cote d'Ivoire Azito CCGT Thermal 3

Cote d'Ivoire Kossou Hydro 3

Ghana Akosombo Hydro 6

Ghana TAPCO Thermal 3

Ghana TICO Thermal 2

Ghana Bui Hydro 3

Ghana Kpong Hydro 4

Ghana TT1PP Thermal 1

Ghana Sunon Asogli I Thermal 6

Ghana Kpone Thermal 3

Mali Manantali Hydro 6

Mali Felou Hydro 3

Nigeria Egbin Thermal 6

Nigeria Kainji Hydro 8

Nigeria Okpai Thermal 3

Nigeria Afam IV Thermal 2

Nigeria Afam V Thermal 2

Senegal Bel Air Thermal 7

Senegal Kaolac Thermal 6

Senegal Cap des Biches C4 Thermal 4

Mali Selingue Hydro 4

Niger Salkadamna Thermal 4

Niger Maradi Thermal 3

Table 46: List of units with PSS installed - 2022

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 166/176

Fina

lver

sion

Small Signal Stability

THEORETICAL BACKGROUND

The majority of power system components such as generators, excitationsystems, governors and load have very nonlinear characteristics. Thesecomponents and their associated controls include saturation and outputlimitations. Despite the fact that the nonlinear systems theory can be used to studysuch a system, this is only valid for small and simple systems, which is not thecase of power systems.

On the other hand, the theory of linear systems can provide useful insight into theoperating behavior of an interconnected power system. However, this theory isonly applicable under the assumption that the dynamic behavior of the system islinear or quasi-linear. Fortunately, low frequency oscillations in a power systemare fairly linear when caused by disturbances of small magnitude such as therandom fluctuation of generation and load. The variations in system dynamicvariables such as machine rotor angle and speed are also small under thesecircumstances and the assumption of a linear system model around an operatingequilibrium point provides valuable results. These conclusions are generallyconsistent with that is observed in the field under similar operating conditions.

The advantage of assuming a linear model for the system is that the theory oflinear systems is in a mature state, which means that methodologies, algorithmsand tools able to deal with very large systems in reasonable computation time areavailable.

In power systems, the study of system stability using linear models is commonlyreferred to as "small-signal stability analysis". This type of study allows theanalysis of the so-called steady-state stability. The following types of oscillationmodes can be detected and identified through small-signal stability analysis:

· Local modes (machine-system modes): associated with the oscillations ofunits at a generating station with respect to the rest of the system (oscillationfrequency typically between 1 Hz and 2 Hz). These oscillations are localizedat one station or a small part of the system.

· Inter-area modes: associated with the swinging of many machines in one partof the system against machines in the other parts (oscillation frequencytypically between 0.1 Hz and 1 Hz). Caused by two or more groups ofelectrically close machines being interconnected by weak a weak transmissionnetwork.

· Control modes: associated with generating units and other controls. Theusual causes of instability of such modes are badly tuned excitation systems,speed governors, HVDC converters and SVCs.

· Torsional modes: associated with the turbine-generator shaft systemrotational components. The usual causes of instability of such modes areinteractions with excitation controls, speed governors, HVDC controls, andseries-capacitor-compensated lines.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 167/176

Fina

lver

sion

It must be emphasized that the small-signal stability is a necessary (but notsufficient) condition for the power system operation. As consequence of not beinga sufficient condition, the results of small-signal stability analyses must beassessed through nonlinear time-domain simulations (electromechanicaltransients simulations).

The following section presents a set of definitions related to linear systems theoryand small-signal stability that will be used for the definition of the methodology tobe adopted in this study.

In the sequel it is presented a brief description on the linear system theory aspectsapplied to power system small-signal stability problem.

Low-frequency electromechanical oscillations usually range from less than 1 Hzto 3 Hz other than those with sub-synchronous resonance (SSR). Multi-machinepower system dynamic behavior in this frequency range is usually represented bya set of nonlinear differential and algebraic equations (DAE) in the form of:

( )

),,(),,(

,,

uzxhyuzxg0uzxfx

===&

Where

· f and g are vectors of differential and algebraic equations· h is a vector of output equations· x, z, u and y are the vectors of state variables, algebraic variables, inputs and

outputs, respectively.

The linearization and elimination of the algebraic variables of this system ofnonlinear DAEs results in a linear system in the form of:

DuCxyBuAxx

+=+=&

Eigenvalues and Oscillation Modes

The eigenvalues (λ) of the state matrix (A) describe the dynamic performance ofthe linearized system. These eigenvalues may be real or complex numbers.Complex eigenvalues always occur in conjugate pairs:

iii jwsl ±=

The eigenvalues of the state matrix correspond to the system modes (oscillationmode, if li is complex). The real part (s) relates to the mode damping and theimaginary part (± jw) relates to the oscillation frequency of the mode.

The relationship between the eigenvalues and the system stability is defined bythe absolute stability criteria, which follows:

"A system is stable according to the absolute stability criteria if all eigenvalues ofthe system are located in the left semi-plane of the complex plane (all eigenvaluesmust have negative real parts)".

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 168/176

Fina

lver

sion

When the system is stable, the analysis of the eigenvalues (computation of theirdamping ratio) indicates the damping level of the system. The requirementthereby is that the oscillations get damped rapidly enough.

The frequency of oscillation of a mode, in Hertz, is given by:

pw2

iif =

The damping ratio of a mode is given by:

22ii

ii

ws

sz

+

-=

Eigenvectors and ModeShapes

In power system literature, the right eigenvector associated to an eigenvalue li isknown as the mode shape of li. The mode shape provides important informationon the participation of an individual ma-chine or a group of machines in a particularmode.

The mode shapes are very useful for the identification of coherent groups ofmachines, as well as for the identification of inter-area modes.

ParticipationFactors

In large power systems, it is important to quantify the role of each generator oneach mode. A method generally employed for this purpose is the calculation ofthe participation factors of each mode. The participation factor is a measure of therelative participation of the k-th state variable on the i-th mode, and vice-versa.

The participation factor of a state variable in a given mode is calculated by meansof combining the left and right eigenvectors associated to that mode. Through thiscombination it is possible to produce a dimensionless measure, which is essentialwhen comparing state variables of different physical units.

Through the calculation of the participation factors it is possible to determine thegenerators that have more contribution to a given oscillation mode. Thegenerators with highest participation factor on poorly damped low frequencymodes are potential candidates for power system stabilizer allocation (PSS).

One drawback of the participation factors is that they are related only to the stateand do not take into account the input/output (I/O) relationship. It cannoteffectively identify a controller site and an optimal feedback signal withoutinformation on the input and output which is more important when output feedbackis employed.

However, the effectiveness of control can be indicated through controllability andobservability factors, as described in the following.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 169/176

Fina

lver

sion

Modal ControllabilityandObservability

Considering the linearized dynamic system given by

CxyBuAxx

=+=&

where

· x is the system state vector,· u is the input vector and· y is the output vector.

Denoting by F the matrix where each column is an eigenvector of A then thechange of variable z = Fx leads to the following system of equations:

xCΦyBuΦΛzz 1

=+= -&

where L is a diagonal matrix composed by the eigenvalues of A. One can seethat:

· If the i-th row of F-1B is not zero, then it is possible to control the i-th modethrough the control u. If there are different potential controls (u is a vector),then the different elements of the i-th row give indication on which inputs havethe largest impacts on the i-th mode.

· If the i-th column of CF is not null, then it is possible to observe the i-th modetrough the output y. If there are different potential observers (y is a vector) thenthe different elements of the i-th column give indication on which outputprovides the more information on the i-th mode.

This means that the controllability of the input signal and the observability of thefeedback signal are basic requirements for PSS allocation.

METHODOLOGY FOR SMALL-SIGNAL STABILITY ANALYSIS

The first step for the small-signal stability analysis is the linearization of the powersystem dynamic model around a steady-state operating point. The linearizedsystem is then used to compute the following quantities:

· System eigenvalues and eigenvectors;· Participation factors;· Controllability indices;· Observability indices.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 170/176

Fina

lver

sion

Identificationof Inter-Area and Critical OscillationModes

Critical oscillation modes are defined as modes with low damping level (a CIGRE"task force"17 about the oscillations in networks recommends a minimum dampingof 5%). The identification of critical oscillation modes starts by the computation ofthe system eigenvalues. As the power system model is very large, thecomputation of all system eigenvalues through orthogonal decomposition-basedmethods (i.e. QR factorization) is not recommended due to the largecomputational time and memory usage required by these methods.

In this project, the method used for eigenvalue calculation is based on thecalculation of all eigenvalues within a predefined region of the complex plane (byemploying an eigenvalue computation algorithm based on the Arnoldi method).This region is defined by the user and must comprise the modes with oscillationfrequency up to 3.0 Hz and damping ratios at least up to 35%. This allows thecalculation of all critical electromechanical modes of the system, as well as theinter-area modes.

The result of eigenvalues computation is a table containing all information relatedto the modes: real and imaginary parts, damping ratio and oscillation frequency.

Critical modes are identified as the ones whose damping ratio is less than 5%.Inter-area modes are pre-identified by selecting the modes whose frequency liein the range between 0.1 Hz and 1.5 Hz. To get the final decision on which modesare in fact inter-area modes, a second step is needed: analysis of the modeshapes, which is explained in the sequel.

ParticipationFactorand ModeShapeAnalysis

In this project, the goal of participation factor and mode shape analysis is toidentify the inter-area modes within the modes with frequency between 0.1 Hzand 1.5 Hz.

Analysis of participation factors:

The participation factors provide an indication of the contribution of the machinesin a given mode. This is very useful for identifying the machines that have majorcontributions to the critical modes as well as to the inter-area modes.

In this study, the participation factors of all modes classified as critical (ζ < 5%) inthe eigenvalue computation phase are calculated and analyzed in order to provideindications on which machines have most participation on the critical modes.

Analysis of mode shapes

As previously described, the mode shapes give the relative magnitude and phaseof the oscillations as seen from a given state variable. Since the objective of thisproject is to analyze electromechanical oscillations, the rotor speed or angle mustbe chosen as state variable.

17 Source: "Analysis and Control of Power System Oscillations", Task force 07, Study Committee 38, December 1996.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 171/176

Fina

lver

sion

In this study, the mode shapes of all oscillation modes with frequency between0.1 Hz and 1.5 Hz are calculated and analyzed in order to identify all inter-areamodes. After the identification of these modes, their respective damping ratios arecarefully analyzed. In case of poorly damped inter-area modes, the necessarymeasures to improve the damping are recommended.

Determination of Candidate Machines for PSS installation orretuning aiming atimproving OscillationDamping

In case of the presence of critical inter-area modes, the identification of thecandidate machines for PSS installation/retuning is performed.

The choice of the machine and the input signals to be used for the improving thedamping of critical modes is not straightforward. It depends on the calculation ofthe controllability and observability indices.

It has to be noticed that the specification and tuning of PSS in order to improvethe damping of critical oscillation modes is out of the scope of this project.

Dynamic Security Assessment – Methodology

The objective of the DSA is to assess the security of the system from a dynamicpoint of view. It can be seen as an evolution of the static security assessment (N-1 criteria).

In this project, the focus of the DSA is on the system stability and voltage recoveryafter incidents occurring at the interconnection lines (tie-lines) and critical linesaffecting cross-border flows. The sizing incident for the DSA is a three-phaseshort-circuit to ground at the terminal of the line cleared in base-time (100ms) by means of tripping the faulted line (opening at both terminals), as depictedin Figure 72.

Figure 72: Sizing incident for DSA analysis.

The acceptance criteria are the following:

· No machine losing synchronism;· No activation of over-/under-voltage, over-/under-frequency relays of the

machines;· Voltage recovery criteria:

- V > 0.70 pu within 500 ms after fault clearance;

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 172/176

Fina

lver

sion

- V > 0.90 pu after 10 seconds.

This analysis was performed for the most critical conditions from system stabilitypoint of view: peak and off-peak load conditions:

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 173/176

Fina

lver

sion

Dynamic Security Assessment – Results

Type ofline Line name Bus 1 Country

of Bus 1 Bus 2 Countryof Bus 2

NominalVoltage

(kV)

Faultat

busResults Peak -dynamic load

Results Off-Peak -

dynamic loadComments

Interco. KAMAKW03-LINSAN03-1 LINSAN03 GU KAMAKW03 SL 225 1 yes yes -Interco. KAMAKW03-LINSAN03-1 KAMAKW03 SL LINSAN03 GU 225 2 yes yes -

Interco. MAN____5-YEKEPA03-2 MAN____5 CI YEKEPA03 LI 225 1 voltage collapseat Ferke (CIV) yes Undamped voltage oscillations due

to the interarea mode leads tovoltage collapse in Côte d’Ivoire.The issue should be solved oncethe interarea is better dampened.

Interco. MAN____5-YEKEPA03-2 YEKEPA03 LI MAN____5 CI 225 2 voltage collapseat Ferke (CIV) yes

Interco. OUAGAD02-NIAMRD02-1 GOROUB02 NR OUAGAE02 BU 330 1 voltage collapse(BU) yes Solved by adding SVC at

Salkadama (NR)

Interco. OUAGAD02-NIAMRD02-1 OUAGAE02 BU GOROUB02 NR 330 2 voltage collapse(BU) yes Solved by adding SVC at

Salkadama (NR)Interco. 1115DUNK-BINGER33-1 BINGER33 CI 1115DUNK GH 330 1 yes yes -Interco. 1115DUNK-BINGER33-1 1115DUNK GH BINGER33 CI 330 2 yes yes -Interco. 30101L_1029VOLT-1 DAWA__02 GH 30101LOM TB 330 1 yes yes -Interco. 30101L_1029VOLT-1 30101LOM TB DAWA__02 GH 330 2 yes yes -Interco. 13003-SAKETE02-1 13003 NI SAKETE02 TB 330 1 yes yes -Interco. 13003-SAKETE02-1 SAKETE02 TB 13003 NI 330 2 yes yes -Interco. BAKEL_03-KAYES_03-1 KAYES_03 MA BAKEL_03 SE 225 1 yes yes -Interco. BAKEL_03-KAYES_03-1 BAKEL_03 SE KAYES_03 MA 225 2 yes yes -

Interco. GAZAOU06-52012-1 52012 NI GAZAOU06 NR 132 1 voltage collapsein NR

voltage collapsein NR Localized voltage collapses at the

end of long radial feeder.Interco. GAZAOU06-52012-1 GAZAOU06 NR 52012 NI 132 2 voltage collapsein NR

voltage collapsein NR

Interco. ZABORI02-33002-1 33002 NI ZABORI02 NR 330 1 yes yes -Interco. ZABORI02-33002-1 ZABORI02 NR 33002 NI 330 2 yes yes -National 4_PA_225-4ZAGT225-1 4_PA_225 BU 4ZAGT225 BU 225 1 yes yes -National 4_PA_225-4ZAGT225-1 4ZAGT225 BU 4_PA_225 BU 225 2 yes yes -National B5PLUS02-DAWA__02-1 B5PLUS02 GH DAWA__02 GH 330 1 yes yes -National B5PLUS02-DAWA__02-1 DAWA__02 GH B5PLUS02 GH 330 2 yes yes -

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 174/176

Fina

lver

sion

Type ofline Line name Bus 1 Country

of Bus 1 Bus 2 Countryof Bus 2

NominalVoltage

(kV)

Faultat

busResults Peak -dynamic load

Results Off-Peak -

dynamic loadComments

National MAGLE_SAK SAKETE02 TB MG2 TB 330 1 yes yes -National MAGLE_SAK SAKETE02 TB MG2 TB 330 2 yes yes -Interco. FERKE__5-829PCRE-1 829PCRE BU FERKE__5 CI 225 1 yes yes -Interco. FERKE__5-829PCRE-1 829PCRE BU FERKE__5 CI 225 2 yes yes -Interco. BOKE__03-SALTHI03-1 BOKE__03 GU SALTHI03 GB 225 1 yes yes -Interco. BOKE__03-SALTHI03-1 SALTHI03 GB BOKE__03 GU 225 2 yes yes -Interco. SAMBAN03-MALI__03-1 MALI__03 GU SAMBAG03 SE 225 1 yes yes -Interco. SAMBAN03-MALI__03-1 SAMBAG03 SE MALI__03 GU 225 2 yes yes -Interco. SIGUIR03-SANAKO03-2 SANAKO03 MA SIGUIR03 GU 225 1 yes yes -Interco. SIGUIR03-SANAKO03-2 SIGUIR03 GU SANAKO03 MA 225 2 yes yes -Interco. TAMBAC03-KAYES_03-1 KAYES_03 MA TAMBAC03 SE 225 1 yes yes -Interco. TAMBAC03-KAYES_03-1 TAMBAC03 SE KAYES_03 MA 225 2 yes yes -Interco. 837PCRE-833PCRE-1 833PCRE MA 837PCRE CI 225 1 yes yes -Interco. 837PCRE-833PCRE-1 837PCRE CI 833PCRE MA 225 2 yes yes -

Table 47: Complete results of the DSA analysis - 2022

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 175/176

Fina

lver

sion

Frequency stability

FREQUENCY STABILITY ANALYSIS - METHODOLOGY

Frequency stability reflects the ability of the system to face unexpected activepower unbalance such as the sudden loss of power infeed or large loads.Frequency stability is usually ensured through the provision of operating reserves(primary and secondary) and under-frequency load-shedding (UFLS) schemes.High probability events are secured via the operating reserves, while lowprobability events are secured with the support of UFLS schemes.

The sizing incident is usually determined to achieve a trade-off between reserveprovision and energy production. Sufficient operating reserves with adequatetechnical performance must be secured to cover the sizing incident withoutresulting in loss of load.

The under-frequency load shedding (UFLS), only activated in case of non-normative incidents, is organized in various steps to minimize the amount of loadshed following frequency transients.

In this study, the frequency stability analysis aims at:

· Assessing the adequacy of the operating reserves for covering the loss of thelargest unit of the interconnected system;

· Identifying the most critical frequency transients;

ALLOCATION OF OPERATING RESERVE

The operating reserve is allocated based on the biggest machine in theinterconnected network. In 2022, this biggest machine is one unit of Eglin 2 whichhas a size of 300 MW. For security purposes, it is a common assumption todispatch this reserve to a value of 110% of this biggest machine.

For 2022, the reserve needs are thus evaluated to be of 330MW at the peak. Thisreserve is allocated to the different countries of the WAPP based on the powergenerated by each machine at the peak such that:

= ∗

Where Rcountry is the reserve allocated to the specific country, Rtotal is the totalneed for primary reserves, Pcountry is the production of the specific country andPtotal is the total production.

WAPP-MP/4NT/0626321/003/03 · Ed. 2019/01/14 176/176

Fina

lver

sion

Country Allocated primary reserve(MW)

Cote d’Ivoire 29

Ghana 64

Senegal 1

The Gambia 0

Guinea 10

Guinee-Bissau 0

Liberia 1

Sierra Leone 1

Mali 5

Burkina 1

Niger 11

Nigeria 206

Togo-Benin 2

TOTAL 330

Table 48: Reserve Allocation - Peak 2022

Once the contribution of each country is determined it is necessary to allocate thisreserve among the different units of the system. To ensure adequate technicalperformance of the primary reserve it is required that this reserve is spread amongdifferent units of the system so that the maximum contribution of a single unit tothe primary reserve should be limited to about 5% of its nominal capacity.

The following assumptions have been considered:

· For the CCs, primary reserve is allocated only on gas turbines since the outputof the linked steam turbine is strictly subject to their operating points;

· PV solar and wind farms do not provide reserve;· Every conventional unit (existing or planned) should have its governor

unblocked and contribute to the operating reserve;

At the helm of the Energy Transition, Tractebel provides a full range of engineering andconsulting services throughout the life cycle of its clients’ projects, including design andproject management. As one of the world’s largest engineering consultancy companies andwith more than 150 years of experience, it's our mission to actively shape the world oftomorrow. With about 4,400 experts and offices in 33 countries, we are able to offer ourcustomers multidisciplinary solutions in energy, water and infrastructure.

TRACTEBEL ENGINEERING S.A.Boulevard Simón Bolívar 34-361000 - Bruxelles - BELGIQUEtractebel-engie.com

Laurence CHARLIERtel. +32 2 476 31 07 [email protected]

update of the ecowas revised master plan for the ... · update of the ecowas revised master plan...

Documents